Mobile devices, access network nodes, and methods

Abstract

The purpose of this disclosure is to provide a method, user equipment, and access network node that can enable a UE to use an AI / ML model more efficiently and reliably. In a first exemplary embodiment, the method performed by user equipment (UE) includes: running an artificial intelligence or machine learning (AI / ML) model in a cell of a first access network node; receiving information from the first access network node for a handover from the first access network node to a second access network node, wherein the information for the handover includes information regarding the continuity of the execution of the AI / ML model; and initiating a process based on the information regarding the continuity of the execution of the AI / ML model.

Description

Technical Field

[0001] The present disclosure relates to a method, a user equipment, and an access network node.

Background Art

[0002] Under the 3rd Generation Partnership Project (3GPP (registered trademark)) standard, NodeB (or eNB in LTE, gNB in 5G) is a Radio Access Network (RAN) node (or simply an "access node", "access network node" or "base station"), through which a communication device (user equipment or "UE") connects to the core network and communicates with other communication devices or remote servers.

Prior Art Documents

Non-Patent Documents

[0003]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] An improved method is needed to propagate artificial intelligence and machine learning (AI / ML) models and related information between nodes in a communication network, and to improve the continuity of AI / ML model use after handover procedures. For example, a UE handover may occur from a source base station to a target base station, and the AI / ML models supported for use in the target base station's cells may not be the same as those supported for use in the source base station's cells. An improved method is needed to enable UEs to use AI / ML models in the target base station's cells more efficiently and reliably after handover.

[0005] One example of the purpose of this disclosure is to provide methods, user equipment, and access network nodes that can enable UEs to use AI / ML models more efficiently and reliably. [Means for solving the problem]

[0006] In the first exemplary embodiment, the method performed by the user equipment (UE) is: The first access network node cell executes an artificial intelligence or machine learning (AI / ML) model, Receiving information for handover from the first access network node to the second access network node, wherein the information for handover includes information regarding the continuity of AI / ML model execution. The process is initiated based on information regarding the continuity of AI / ML model execution, Includes.

[0007] In a second exemplary embodiment, the method performed by the first access network node is: The handover information from the first access network node to the second access network node is transmitted to user equipment (UE) running an artificial intelligence or machine learning (AI / ML) model in a cell of the first access network node, wherein the transmitted handover information includes information regarding the continuity of AI / ML model execution. Includes, Information regarding the continuity of AI / ML model execution prompts the UE to initiate the process for ensuring the continuity of AI / ML model execution.

[0008] In a third exemplary embodiment, the method performed by the second access network node is: The first access network node transmits information for handover from the first access network node to the second access network node, wherein the information for handover includes information regarding the continuity of execution of the artificial intelligence or machine learning (AI / ML) model by user equipment (UE) in the cell of the first access network node. Includes, Information regarding the continuity of AI / ML model execution is sent to the UE, prompting it to initiate the process for ensuring the continuity of AI / ML model execution.

[0009] In a fourth exemplary embodiment, user equipment (UE) is: A means for executing an artificial intelligence or machine learning (AI / ML) model in a cell of the first access network node, A means for receiving information for handover from a first access network node to a second access network node, wherein the handover information includes information regarding the continuity of execution of the AI / ML model. A means of initiating a process based on information regarding the continuity of AI / ML model execution, Includes.

[0010] In a fifth exemplary embodiment, the first access network node is: Means for transmitting information for handover from a first access network node to a second access network node to user equipment (UE) running an artificial intelligence or machine learning (AI / ML) model in a cell of a first access network node, wherein the information for handover includes information regarding the continuity of execution of the AI / ML model. Includes, Information regarding the continuity of AI / ML model execution prompts the UE to initiate the process for ensuring the continuity of AI / ML model execution.

[0011] In a sixth exemplary embodiment, the second access network node is: Means for transmitting information for a handover from a first access network node to a second access network node, wherein the information for the handover includes information regarding the continuity of execution of an artificial intelligence or machine learning (AI / ML) model by user equipment (UE) in a cell of the first access network node. Includes, Information regarding the continuity of AI / ML model execution is sent to the UE, prompting it to initiate the process for ensuring the continuity of AI / ML model execution. [Effects of the Invention]

[0012] This disclosure provides methods, user equipment, and access network nodes that can enable UEs to use AI / ML models more efficiently and reliably.

Brief Description of the Drawings

[0013] Exemplary embodiments of the present disclosure are now described by way of example with reference to the accompanying drawings. [Figure 1] FIG. 1 is a schematic diagram of a mobile (“cellular” or “wireless”) electrical communication system. [Figure 2] FIG. 2 is a diagram showing a typical frame structure that can be used in the electrical communication system of FIG. 1. [Figure 3] FIG. 3 is a schematic block diagram showing the main components of a distributed unit (DU) 50 that can be used as part of the RAN device 5 for the communication system 1 shown in FIG. 1. [Figure 4] FIG. 4 is a schematic block diagram showing the main components of a central unit (CU) 60 that can be used as part of the RAN device 5 for the communication system 1 shown in FIG. 1. [Figure 5] FIG. 5 shows a mobility procedure in which a handover occurs from a source (R)AN node to a target (R)AN node. [Figure 6] FIG. 6 shows a random access (RA) procedure that can be executed in the system of FIG. 1. [Figure 7] FIG. 7 shows a schematic diagram of point-to-point transmission and point-to-multipoint transmission. [Figure 8] FIG. 8 is a diagram showing a framework related to an AI / ML model. [Figure 9] FIG. 9 is a diagram showing a method of training an AI / ML model and monitoring the performance of the AI / ML model. [Figure 10] FIG. 10 is a diagram showing an example of an AI / ML request and an AI / ML response. [Figure 11] FIG. 11 is a diagram showing an example of AI / ML information update. [Figure 12] FIG. 12 is a diagram showing an example of a method for broadcasting a notification of an AI / ML model supported by a base station. [Figure 13] Figure 13 shows an example of an AI / ML model being transmitted from an AI / ML server to a UE via a base station. [Figure 14] Figure 14 shows an example where an AI / ML model is transmitted from a CU (Unit) of a distributed base station to an UE (User Environment) via a DU (Unit). [Figure 15] Figure 15 shows an example of the AI / ML model function area. [Figure 16] Figure 16 shows how AI / ML model area information is received by the UE. [Figure 17] Figure 17 is a modified version of Figure 5, in which various steps of the method have been changed to include the transmission of an AI / ML model and / or AI / ML model-related information. [Figure 18] Figure 18 is a schematic block diagram showing the main components of the UE for the telecommunications system in Figure 1. [Figure 19] Figure 19 is a schematic block diagram showing the main components of a base station for the telecommunications system shown in Figure 1. [Figure 20] Figure 20 is a schematic block diagram showing the main components of the core network nodes or functions for the telecommunications system in Figure 1. [Modes for carrying out the invention]

[0014] This disclosure relates to communication systems. This disclosure has a non-exclusive but specific relevance to wireless communication systems and devices operating in accordance with 3GPP standards or equivalent standards (including LTE Advanced, Next Generation or 5G networks, Future Generation and beyond) or derivative standards. This disclosure is not necessarily exclusive but is particularly relevant to AI / ML models used in “new wireless” systems (also called “next generation” systems) and similar systems.

[0015] (Related technologies) Recent developments in 3GPP standards are referred to as Long-Term Evolution (LTE) and Evolved UMTS Terrestrial Radio Access Network (E-UTRAN) of Evolved Packet Core (EPC) networks, commonly known as "4G." The terms "5G" and "new radio" (NR) refer to evolving communication technologies expected to support a variety of applications and services. Various details of 5G networks are described in the "NGMN 5G White Paper" V1.0 by the Next Generation Mobile Network (NGMN) Alliance, which can be found, for example, at https: / / www.ngmn.org / 5g-white-paper.html. 3GPP intends to support 5G through the so-called 3GPP Next Generation (NextGen) Radio Access Network (RAN) and 3GPP NextGen core networks.

[0016] Under the 3GPP standard, a NodeB (or eNB in ​​LTE, gNB in ​​5G) is a Radio Access Network (RAN) node (or simply an “access node,” “access network node,” or “base station”) through which communication devices (user equipment or “UE”) connect to the core network and communicate with other communication devices or remote servers. For simplicity, this application uses the terms RAN node, base station, or access network node to refer to any such access node.

[0017] Some of the additional developments in 3GPP concern the use of artificial intelligence (AI) and machine learning (ML), often abbreviated as AI / ML. Predictions or inferences generated using AI / ML models can be used as part of various methods to improve the reliability or efficiency of communications within a network. For example, an AI / ML model can be used to predict the path of a UE based on the UE's previous mobility, can be used for beam management, or can be used in methods of encoding and transmitting information. The AI / ML model may be hosted at a base station, which may perform control over communication resources or control over the state of a UE (e.g., control of UE mobility, or control of the UE's radio resource, RRC, state control) based on inferences (e.g., decisions or predictions) generated using the AI / ML model. The base station may also transmit the inferences generated using the model to another node in the network for use by other nodes.

[0018] Alternatively, the AI / ML model may be hosted on two nodes in the network, for example, a base station and an UE. In this case, both the base station and the UE can use the model to make decisions or predictions. For example, the UE can use the model as part of an encoding process to encode (and / or compress) channel state information (CSI) for transmission to the base station, and the base station can use the same model as part of a corresponding decoding (and / or decompression) process to decode the CSI received from the UE.

[0019] (Challenges related to related technologies) An improved method is needed to propagate AI / ML models and related information between nodes in a communication network and to improve the continuity of AI / ML model use after handover procedures. For example, a UE handover may occur from a source base station to a target base station, and the AI / ML models supported for use in the target base station's cells may not be the same as those supported for use in the source base station's cells. An improved method is needed to enable UEs to use AI / ML models in the target base station's cells more efficiently and reliably after handover.

[0020] Another issue is that the transmission of AI / ML models (e.g., from the source base station) to the UE may be interrupted by the handover procedure. Methods are needed to mitigate such interruptions.

[0021] More generally, there is a need for improved methods to enable more efficient and reliable transmission of AI / ML models and related information between nodes within a communication network.

[0022] (Description of the manner) This disclosure describes multiple embodiments and variations of each example. These embodiments and variations can be combined in any way.

[0023] In a first aspect, the Disclosure provides a method performed by a first access network node, the method including determining that user equipment (UE) should be handed over from the first access network node to a second access network node; sending a handover request to the second access network node for the handover of the UE to the second access network node; and receiving model information from the second access network node, indicating one or more models or one or more parameters for use with the models, for use by the UE after the handover to the second access network node to determine, predict, or generate output parameters.

[0024] One or more models may be artificial intelligence or machine learning (AI / ML) models.

[0025] Model information can be received from the second access network node in a handover request acknowledgment message.

[0026] Model information can be received from the second access network node in setup or update messages for data transfer via the interface between the first and second access network nodes.

[0027] Model information may be received from a second access network node in a dedicated information element.

[0028] The method may further include determining, based on model information received from a second access network node, which model is to be used by the UE after the UE is handed over to the second access network node; sending the UE at least one of the following to obtain the identity of the model to be used by the UE after the handover to the second access network node, the model to be used by the UE after the handover to the second access network node, or information that the UE will use to obtain the model to be used by the UE after the handover to the second access network node; sending a notification to the second access network node that the second access network node should send the model to the UE; or sending a notification to a server or core network node that the server or core network node should send the model to the UE.

[0029] This method may include sending a model for use by the UE after handover in an RRC reconstruction message, or sending information for the UE to use to retrieve the model in an RRC reconstruction message.

[0030] This method may include receiving a model from a second access network node for use by the UE after the UE is handed over to the second access network node, and sending the model to the UE, or receiving information from the second access network node for the UE to use in obtaining a model for use by the UE after the handover to the second access network node, and sending the information to the UE for the UE to use in obtaining the model.

[0031] The information that the UE uses to retrieve the model may include the information that the UE uses to retrieve the model from a server or core network node.

[0032] This method may include sending a notification to the second access network node of one or more models available for use by the UE before handover to the second access network node for making decisions, predictions, or generating output parameters.

[0033] Notifications of one or more models available for use by the UE before handover may include notifications of one or more models supported for a particular use case or feature.

[0034] Notifications of one or more models available for use by the UE before handover may include a model identity or model version number.

[0035] Notifications of one or more models available for use by the UE before handover may include notifications that a cell on the first access network node is part of an area associated with each set of one or more models for decision, prediction, or output generation.

[0036] Notification of one or more models available for use by the UE prior to the handover may be included in the handover request sent from the first access network node to the second access network node.

[0037] This method may include sending a notification to the second access network node of one or more models available for use by the UE before a handover in a setup message or update message, in order to transfer data over the interface between the first and second access network nodes.

[0038] This method may include sending a dedicated information element to a second access network node to indicate one or more models available for use by the UE before handover.

[0039] Model information indicating one or more models for use by the UE after handover, or one or more parameters for use with the models, may include at least one of the following: a notification of the use case or functionality of the models supported by the second access network node; the model identity or version number of one or more models for use by the UE after handover; or a notification that a cell on the second access network node is part of an area associated with each set of models for decision, prediction, or output generation.

[0040] Model information indicating one or more models to be used by the UE after handover may include notifications of multiple models that the UE can use for a particular use case or function after handover.

[0041] This method may further include sending a notification to the UE that, after the handover to the second access network node, the UE should continue using the model for a particular use case or function.

[0042] The method may further include receiving a request from a second access network node to the first access network node to send a model to the second access network node for use by the UE after a handover to the second access network node for decision, prediction, or generation of output parameters, and sending the requested model to the second access network node.

[0043] The method may further include determining, based on model information received from a second access network node, that a model should not be used by the UE or should be disabled, and sending a notification to the UE that the model should not be used by the UE or should be disabled.

[0044] Model information received from the second access network node may include information indicating one or more parameters for use by the UE with the model after the UE is handed over to the second access network node, and the method may further include sending a notification to the UE of one or more parameters for use by the UE with the model after the handover, or sending one or more parameters to the UE for use by the UE with the model after the handover.

[0045] Notifying one or more parameters, or sending one or more parameters to the UE, may include notifying one or more parameters, or sending one or more parameters to the UE in an RRC reconfiguration message.

[0046] In another aspect, the Disclosure provides a method performed by a second access network node, the method including receiving a handover request from the first access network node for the handover of a UE from the first access network node to the second access network node, and transmitting to the first access network node model information indicating one or more models, or one or more parameters for use with a model, for use by the UE after the handover to the second access network node to generate determination, prediction, or output parameters.

[0047] One or more models may be artificial intelligence or machine learning (AI / ML) models.

[0048] Sending model information may include sending model information to the first access network node in a handover request acknowledgment message.

[0049] Transmitting model information may include sending the model information to the first access network node in a setup message or update message in order to transfer data over the interface between the first and second access network nodes.

[0050] Sending model information may include sending model information to a first access network node using a dedicated information element.

[0051] The method may further include receiving a notification from the first access network node that the second access network node should send a model to the UE or send information to the UE to retrieve the model, and the model is to be used by the UE after the UE is handed over to the second access network node.

[0052] This method may include sending a model to the first access network node for use by the UE after the UE is handed over to the second access network node, or sending information to the first access network node for use by the UE or for use by the first access network node in order to obtain a model to be used by the UE after the handover to the second access network node.

[0053] The information used by the UE or the first access network node to retrieve the model may include information for retrieving the model from the server or core network node.

[0054] This method may include receiving notifications from the first access network node of one or more models available for use by the UE before handover to a second access network node for making decisions, predictions, or generating output parameters.

[0055] This method may further include determining which models the UE will use after the handover, based on a notification indicating which models are available for the UE to use before the handover.

[0056] Model information indicating one or more models for use by the UE after handover, or one or more parameters for use with the models, may include at least one of the following: a notification of the use case or functionality of the models supported by the second access network node; the model identity or version number of one or more models for use by the UE after handover; or a notification that a cell on the second access network node is part of an area associated with each set of models for decision, prediction, or output generation.

[0057] Model information can include notifications of multiple models that the UE can use for a specific use case or function after handover.

[0058] This method may further include receiving notifications from the UE about multiple models that may be used by the UE for specific use cases or functions that should be used by the UE after the handover.

[0059] The method may further include sending a request to the first access network node for sending a model to the second access network node for use by the UE after a handover to the second access network node for making decisions, predictions, or generating output parameters, and receiving the requested model from the first access network node.

[0060] The model information transmitted to the first access network node may include information indicating one or more parameters for the UE to use with the model after the UE is handed over to the second access network node.

[0061] In another aspect, the Disclosure provides a method performed by user equipment (UE) which includes receiving model information from a first access network node indicating a model to be used by the UE after a handover of the UE to a second access network node for determining, predicting, or generating output parameters, wherein the model information includes at least one of the following: notification of the identity of the model to be used by the UE after a handover of the UE to a second access network node; one or more parameters of the model to be used by the UE after a handover to a second access network node; the model to be used by the UE after a handover to a second access network node; or information to be used by the UE to obtain the model; and performing a handover procedure for the handover of the UE from the first access network node to a second access network node.

[0062] The model may be an artificial intelligence or machine learning (AI / ML) model.

[0063] If the model information includes information used by the UE to retrieve the model, this method may further include retrieving the model.

[0064] Obtaining a model can include retrieving the model from a server, a core network node, or a second access network node.

[0065] This method may further include using a model to make decisions, predictions, or generate output parameters after the UE handover from the first access network node to the second access network node.

[0066] This method may further include configuring the model for use by the UE before handing over the UE to the second access network node.

[0067] Model information may include notification of multiple models that the UE can use for a particular use case or function after handover, and the method may further include determining which of the multiple models to use for the use case or function, and sending a notification to a second access network node that the UE will use the determined model for the use case or function, or a request to the UE to use the determined model for the use case or function.

[0068] The method may further include receiving notification from the first access network node that the UE should continue to use a model for a particular use case or function after the handover to the second access network node, and continuing to use the model for the use case or function after the handover to the second access network node.

[0069] This method may further include receiving a notification from the first access network node that the model should be disabled, and disabling the use of the model in the UE before the handover to the second access network node.

[0070] This method may include receiving information from a first access network node indicating one or more parameters for the UE to use with the model after the UE is handed over to a second access network node, and using one or more parameters with the model after the UE is handed over to the second access network node.

[0071] This method may further include determining, before the handover, whether to maintain the model in the UE's memory during the handover of the UE to a second access network node.

[0072] The method may further include transmitting to the second access network node the model to be used by the UE after the UE is handed over to the second access network node, or information to be used by the second access network node to obtain the model to be used by the UE after the UE is handed over to the second access network node.

[0073] In another aspect, the Disclosure provides a first access network node, the first access network node comprising means for determining that user equipment (UE) should be handed over from the first access network node to a second access network node; means for transmitting a handover request to the second access network node for the handover of the UE to the second access network node; and means for receiving from the second access network node model information indicating one or more models, or one or more parameters for use with a model, for use by the UE after the handover to the second access network node for determination, prediction, or generating output parameters.

[0074] In another aspect, the Disclosure provides a second access network node, the second access network node, which includes means for receiving a handover request from the first access network node for the handover of a UE from the first access network node to the second access network node, and means for transmitting to the first access network node model information indicating one or more models, or one or more parameters for use with a model, for use by the UE after the handover to the second access network node, for the purpose of generating determination, prediction, or output parameters.

[0075] In another aspect, the Disclosure provides user equipment (UE) comprising: means for receiving model information from a first access network node indicating a model to be used by the UE after a handover of the UE to a second access network node for determining, predicting, or generating output parameters, wherein the model information includes at least one of the following: notification of the identity of a model to be used by the UE after a handover of the UE to a second access network node; one or more parameters of a model to be used by the UE after a handover to a second access network node; a model to be used by the UE after a handover to a second access network node; or information to be used by the UE to obtain a model; and means for performing a handover procedure for a handover of the UE from a first access network node to a second access network node.

[0076] (overview) Here, for illustrative purposes only, we will describe an exemplary communication system in general terms, referring to Figures 1 and 2.

[0077] Figure 1 is a schematic diagram of a mobile ("cellular" or "wireless") communication system 1, to which exemplary embodiments of the present disclosure are applicable.

[0078] In communication system 1, user equipment (UE) 3-1, 3-2, 3-3 (e.g., mobile phones and / or other mobile devices) can communicate with each other via Radio Access Network (RAN) nodes 5 (base stations 5, RAN equipment 5) operating according to one or more compatible radio access technologies (RATs). In the illustrated example, RAN node 5 comprises an NR / 5G base station or "gNB" 5 operating one or more associated cells 9. Communication via base station 5 is typically routed via a core network 7 (e.g., a 5G core network or an evolved packet core network (EPC)).

[0079] As those skilled in the art will understand, three UE3 and one base station 5 are shown in Figure 1 for illustrative purposes, but the system, when implemented, typically includes other base stations 5 and UE3.

[0080] Each base station 5 controls one or more associated cells 9 directly or indirectly through one or more other nodes (e.g., home base stations, repeaters, remote radio heads, distributed units, etc.). It will be understood that base stations 5 may be configured to support 4G, 5G, 6G, and / or any other 3GPP or non-3GPP communication protocols.

[0081] The UE3s and their serving base stations 5 are connected via appropriate air interfaces (such as the so-called "Uu" interface). Neighboring base stations 5 may be connected to each other via appropriate inter-base station interfaces (such as the so-called "X2" interface, "Xn" interface, etc.).

[0082] The core network 7 includes several logical nodes (or “functions”) to support communication in the communication system 1. In this example, the core network 7 comprises a control plane function (CPF) 10 and one or more user plane functions (UPF) 11. The CPF 10 includes one or more Access and Mobility Management Functions (AMFs) 10-1, one or more Session Management Functions (SMFs), and several other functions 10-n.

[0083] Base station 5 is connected to the core network nodes via appropriate interfaces (or "reference points"), such as an N2 reference point between base station 5 and AMF10-1 for control signaling communications, and an N3 reference point between base station 5 and each UPF11 for user data communications. Each UE3 is connected to AMF10-1 via a logical non-access stratum (NAS) connection on the N1 reference point (similar to the S1 reference point in LTE). It will be understood that N1 communications are routed transparently through base station 5.

[0084] One or more UPF11s are connected to an external data network 20 (such as an IP network like the Internet) via a reference point N6 for the communication of user data.

[0085] The AMF10-1 performs mobility management functions, maintains NAS signaling connections with each UE3, and manages UE registration. The AMF10-1 also manages paging. The SMF10-2 provides session management functions (forming part of the MME function in LTE) and also incorporates several control plane functions (provided by the serving gateway and packet data network gateway in LTE). The SMF10-2 also assigns IP addresses to each UE3.

[0086] The base station 5 of communication system 1 is configured to operate at least one cell 9 on an associated TDD carrier operating in a non-paired spectrum. It will be understood that base station 5 may also operate at least one cell 9 on an associated FDD carrier operating in a paired spectrum.

[0087] Base station 5 is also configured to transmit control information and user data via several downlink (DL) physical channels and to transmit several physical signals, and UE3 is configured to receive control information and user data via several DL physical channels and to transmit several physical signals. DL physical channels correspond to resource elements (REs) that carry information transmitted from higher layers, and DL physical signals correspond to REs used in the physical layer that do not carry information transmitted from higher layers.

[0088] Physical channels may include, for example, a physical downlink shared channel (PDSCH), a physical broadcast channel (PBCH), and a physical downlink control channel (PDCCH). The PDSCH carries data that shares the PDSCH's capacity on a time and frequency basis. The PDSCH can carry various data items, including, for example, user data, UE-specific upper-layer control messages mapped from higher channels, system information blocks (SIBs), and paging. The PDCCH carries downlink control information (DCI) to support several functions, including, for example, scheduling downlink transmissions on the PDSCH and uplink data transmissions on the physical uplink shared channel (PUSCH). The PBCH provides the UE3 with a Master Information Block, or MIB. The PBCH also works in conjunction with the PDCCH to support time and frequency synchronization, which helps with cell acquisition, selection, and re-selection.

[0089] UE3 may receive Synchronization Signal Blocks (SSBs), and UE3 may assume that opportunities to receive PBCHs, primary synchronization signals (PSSs), and secondary synchronization signals (SSSs) are within a consecutive symbol, forming an SS / PBCH block. Base station 5 may transmit several synchronization signal (SS) blocks corresponding to different DL beams. The total number of SS blocks may be limited, for example, to a duration of 5 ms as an SS burst. The period of SSB transmission may be notified to the UE using any appropriate signaling (e.g., per serving cell using ssb-periodicityServingCell). The period value of an SSB may be, for example, 20 ms or more. In the case of initial cell selection, UE3 may be configured to assume that SS bursts occur with a period of 2 frames. UE3 may also be provided with notification of which SSBs will be transmitted within a 5 ms duration (e.g., using ssb-PositionsInBurst).

[0090] DL physical signals may include, for example, a reference signal (RS) and a synchronization signal (SS). The reference signal (sometimes known as a pilot signal) is a signal with a predefined special waveform known to both the UE3 and the base station 5. The reference signal may include, for example, a cell-specific reference signal, a UE-specific reference signal (UE-RS), a downlink demodulation signal (DMRS), and a channel state information reference signal (CSI-RS).

[0091] Similarly, UE3 is configured to transmit control information and user data via several uplink (UL) physical channels corresponding to REs that carry information transmitted from higher layers, and UL physical signals used in the physical layer that do not carry information transmitted from higher layers, and base station 5 is configured to receive control information and user data via several UL physical channels corresponding to REs that carry information transmitted from higher layers, and UL physical signals used in the physical layer that do not carry information transmitted from higher layers. The physical channels may include, for example, PUSCH, physical uplink control channel (PUCCH), and / or physical random-access channel (PRACH). The UL physical signals may include, for example, demodulation reference signal (DMRS) for UL control / data signals, and / or sounding reference signal (SRS) used for UL channel measurement.

[0092] When UE3 first establishes a radio resource control (RRC) connection with base station 5 via cell 9, UE3 registers with the appropriate core network node (e.g., AMF, MME). UE3 is in a so-called RRC connected state, and the associated UE context is maintained by the network. When UE3 is in a so-called RRC idle or RRC inactive state, UE3 selects an appropriate cell for camping so that the network knows UE3's approximate location (though not necessarily at the cell level).

[0093] A base station 5 may be a base station 5 divided between one or more distributed units (DUs) 50 and a central unit (CU) 60, where the CU 60 typically performs higher-level functions and communication with the next-generation core, and the DU 50 performs lower-level functions and communication via an air interface with neighboring UEs 3 (i.e., within a cell operated by a gNB 5). This type of base station 5 may be referred to as a “distributed” base station 5 or gNB 5. A distributed gNB 5 includes the following functional units: gNB Central Unit (gNB-CU): A logical node that controls the operation of one or more gNB-DUs and hosts the gNB's Radio Resource Control (RRC) layer, Service Data Adaptation Protocol (SDAP) layer, and Packet Data Convergence Protocol (PDCP) layer (or the en-gNB's RRC and PDCP layers). The gNB-CU terminates the so-called F1 interface connected to the gNB-DUs. gNB Distributed Unit (gNB-DU): A logical node that hosts the Radio Link Control (RLC) layer, Medium Access Control (MAC) layer, and Physical (PHY) layer of a gNB or en-gNB, and whose operation is partially controlled by a gNB-CU. A single gNB-DU supports one or more cells. A single cell is supported by only one gNB-DU. A gNB-DU terminates the F1 interface connected to the gNB-CU. gNB-CU-Control Plane (gNB-CU-CP): A logical node that hosts the control plane portion of the RRC and PDCP protocols for the gNB-CU for en-gNB or gNB. The gNB-CU-CP terminates the so-called E1 interface connected to the gNB-CU-UP, and the F1-C (F1 control plane) interface connected to the gNB-DU. gNB-CU-User Plane (gNB-CU-UP): A logical node that hosts the user plane portion of the PDCP protocol for the gNB-CU for en-gNB, and the user plane portions of the PDCP protocol and SDAP protocol for the gNB-CU for gNB. gNB-CU-UP terminates the E1 interface connected to gNB-CU-CP and the F1-U (F1 user plane) interface connected to gNB-DU.

[0094] When a distributed base station or a similar control plane-user plane (CP-UP) partition is employed, it will be understood that each of the control plane entity and the user plane entity may include associated transceiver circuits, antennas, network interfaces, control units, memory, operating systems, and communication control modules. When base station 5 comprises a distributed base station, the network interfaces also include E1 and F1 interfaces (F1-C in the case of the control plane and F1-U in the case of the user plane) for communicating signals between the respective functions of the distributed base station.

[0095] (Frame structure) Referring to Figure 2, which shows a typical frame structure that may be used in communication system 1, the base station 5 and UE3 of communication system 1 communicate with each other in the time domain using resources organized into frames of length 10 ms. Each frame consists of 10 equally sized subframes of length 1 ms. Each subframe is divided into one or more slots containing 14 Orthogonal frequency-division multiplexing (OFDM) symbols of equal length.

[0096] As shown in Figure 2, communication system 1 supports several different numerologies (subcarrier spacing (SCS), slot length, and consequently OFDM symbol length). Specifically, each numerology is identified by the parameter μ, where μ=0 represents 15kHz (corresponding to LTE SCS). Currently, SCS for other values ​​of μ can actually be derived from μ=0 by scaling up by a power of 2 (i.e., SCS = 15 × 2). μ (kHz). The relationship between the parameter μ and SCS(Δf) is shown in Table 1. [Table 1] Table 1 shows an example of 5G numerology.

[0097] (RAN equipment) <du> Figure 3 is a schematic block diagram showing the main components of DU50, which may be used as part of RAN equipment 5 for the communication system 1 shown in Figure 1. As shown, DU50 has a radio unit (RU) and a transceiver circuit 451 for transmitting signals to and receiving signals from a communication device (such as UE3) via an associated DU-RU interface 453, and for transmitting signals to and receiving signals from a CU60 of RAN equipment 5 via a CU interface 454 (which has an F1 interface that can be divided into F1-U and F1-C interfaces for user plane and control plane signaling, respectively).

[0098] DU50 has a controller 457 for controlling the operation of DU50. Controller 457 is associated with memory 459. Software may be pre-installed in memory 459 and / or may be downloaded, for example, via communication system 1 or from a removable data storage device (RMD). In this example, controller 457 is configured to control the overall operation of DU50 by program instructions or software instructions stored in memory 459.

[0099] As shown in the figure, these software instructions include, among other things, the operating system 461, the communication control module 463, the F1 module 465, the DU-RU module 468, the DU management module 472, the UE profile management module 473, and the mobility module 475.

[0100] The communication control module 463 is operable to control communication between DU50 and one or more RUs (and thus between DU50 and UE3), and between DU50 and CU60. The communication control module 463 is configured to have overall control over the reception of signals corresponding to uplink communication from UE3 and to handle the transmission of downlink communication to UE3.

[0101] The F1 module 465 is responsible for the proper processing of signals received from or transmitted to the CU60 via one or more CU (e.g., F1) interfaces 454. These signals can be separated into user-plane signals received from or transmitted to the CU-UP portion of the CU60 via the F1-U interface, and control-plane signals received from or transmitted to the CU-CP portion of the CU60 via the F1-C interface.

[0102] The DU-RU module 468 is responsible for the proper processing of signals received from or transmitted to one or more RUs (e.g., DU-RU) interfaces 453.

[0103] The DU management module 472 is responsible for managing the overall operation of the DU50 and the overall performance of the tasks required of the DU50. These tasks include, among other things, generating and transmitting appropriate messages using the appropriate signaling application protocol, depending on the functional division between the RU, DU50, and CU60, such as generating MAC signaling for the interpretation and transmission of received MAC signaling. Where appropriate, the DU management module 472 can control the overall operation of the DU50 in one of the methods described below.

[0104] The UE profile management module 473 is responsible for performing UE profile-related functions, including (where applicable): receiving and storing UE profiles or associated support / preference information from UE3 or other locations in the network; (where applicable) determining appropriate mobility-specific configurations based on UE profile / support / preference information for implementation on UE3 and / or RAN equipment, and / or (where applicable) providing configuration information for appropriately configuring the UE in a mobility-based configuration. The UE profile management module 473 can also store previous mobility information for UE3 (e.g., previous movements of UE3 between different communication cells in the network). Depending on the implementation, it will be understood that the gNB-DU may not have to implement at least some of these features.

[0105] The mobility module 475 is responsible for controlling mobility procedures for one or more UE3s. For example, the mobility module 475 may be configured to perform one or more measurements of UE3 mobility or to select candidate cells for handover.

[0106] <cu> Figure 4 is a schematic block diagram showing the main components of the CU60 RAN equipment for the communication system 1 shown in Figure 1. As shown, the CU60 has transceiver circuits 551 for transmitting signals to and receiving signals from the DU50 via one or more DU interfaces 554 (which include, for example, an F1 interface that can be divided into F1-U and F1-C interfaces for user plane and control plane signaling, respectively), and for transmitting signals to and receiving signals from the functions of the core network 7 via one or more CU interfaces 555 (which include, for example, N2 and N3 interfaces, etc.).

[0107] CU60 has a controller 557 for controlling the operation of CU60. Controller 557 is associated with memory 559. Software may be pre-installed in memory 559 and / or may be downloaded, for example, via communication system 1 or from a removable data storage device (RMD). In this example, controller 557 is configured to control the overall operation of CU60 by program instructions or software instructions stored in memory 559.

[0108] As shown in the figure, these software instructions include, among other things, the operating system 561, the communication control module 563, the F1 module 565, the E1 module 566, the N2 module 568, the N3 module 569, the CU-UP management module 571, the CU-CP management module 572, the UE profile management module 573, and the mobility module 575. The function of the mobility module 575 was previously described with reference to Figure 3.

[0109] The communication control module 563 is operable to control communication between CU60 and one or more DU50 (and thus between CU60 and UE3), and between CU60 and the core network 7. The communication control module 563 is configured to have overall control over the reception of signals corresponding to uplink communication from UE3 and to control the transmission of downlink communication.

[0110] The F1 module 565 is responsible for the proper processing of signals received from or transmitted to the DU 50 via one or more DU (e.g., F1) interfaces 554. These signals include user-plane signals received or transmitted by the CU-UP portion of the CU 60 via the F1-U interface, and control-plane signals received or transmitted by the CU-CP portion of the CU 60 via the F1-C interface.

[0111] The E1 module 566 is responsible for the proper processing of signals transmitted between the CU-UP portion and the CU-CP portion of the CU60 via the corresponding internal CU interface (such as E1).

[0112] The N2 module 568 is responsible for the proper processing of signals received from or transmitted to the AMF8-1 via one or more corresponding CU interfaces 555 (such as N2).

[0113] The N3 module 569 is responsible for the proper processing of signals received from or transmitted to one or more core network user plane functions via one or more corresponding CU interfaces 555 (such as N3).

[0114] The CU-UP management module 571 is responsible for managing the overall operation of the CU-UP portion of the CU60 and the overall performance of the tasks required for CU-UP.

[0115] The CU-CP management module 572 is responsible for managing the overall operation of the CU-CP portion of the CU60 and the overall performance of the tasks required of the CU-CP. These tasks include, among other things, the generation and transmission of appropriate messages using appropriate signaling application protocols, depending on the functional division between the RU, DU50, and CU60, such as the generation of RRC signaling for the interpretation and transmission of received RRC signaling.

[0116] The UE profile management module 573 is responsible for performing functions related to the UE (mobility) profile, including (where applicable): receiving and storing the UE profile or associated support / preference information from the UE3 or other locations in the network; determining an appropriate mobility-specific configuration based on the UE profile / support / preference information for implementation in the UE3 and / or RAN equipment 5, and / or providing configuration information for appropriately configuring the UE in a mobility-based configuration. The UE profile management module 573 can also store previous mobility information for the UE3 (e.g., previous movements of the UE3 between different communication cells in the network). Depending on the implementation, it will be understood that the gNB-CU60 may not have to implement at least some of these features.

[0117] <System Information and SIB> It will be understood that transmissions in cell 9 of base station 5 may include one or more broadcast transmissions, one or more unicast transmissions for reception by UE3, and / or one or more multicast transmissions for reception by a group of UE3. System information (SI) transmitted in a cell may include "minimum SI" (MSI) and "other SI" (OSI). OSI may be broadcast on demand, for example, using a downlink shared channel (DL-SCH). OSI may be broadcast upon request from a UE3 that is in a radio resource control (RRC) idle or RRC inactive state. OSI may also be requested by a UE3 that is in an RRC connected state, for example, via one or more dedicated RRC transmissions.

[0118] The System Information Block (SI) may include information that enables UE3 to complete cell selection (for example, by configuring it to complete it), information that enables UE3 to complete a cell re-selection procedure, or information that enables UE3 to receive one or more paging messages transmitted within a cell. The SI may be broadcast using a Master Information Block (MIB) and one or more System Information Blocks (SIBs).

[0119] An MSI comprises an MIB and a system information block 1 (SIB1). The MIB contains information that the UE3 uses to receive SIB1, such as the subcarrier interval of SIB1. The MIB provides information corresponding to the Control Resource Set (CORESET) and the search space. SIB1 may be called the “remaining MSI” (RMSI). SIB1 may be transmitted in a dedicated RRC message, and other SIBs (such as SIB2-SIB9) may be transmitted using one or more other appropriate RRC transmissions (such as another dedicated RRC message).

[0120] MIB and SIB1 may provide UE3 with scheduling information for receiving and decoding other SIBs such as SIB2-SIB9, and may provide information for UE3 to use to receive one or more paging messages. OSI may include, for example, SIB2-SIB9 transmitted using DL-SCH in SI messages. Mapping of SIB2-SIB9 to corresponding SI messages may be provided to UE3 by base station 5. MIB and SIB1-SIB9 are described in more detail, for example, 3GPP Technical Specification (TS) 38.331. SIB2 provides information for intra-frequency, inter-frequency, and inter-system cell reselection. SIB3 provides cell-specific information for intra-frequency cell reselection. SIB4 provides information for inter-frequency cell reselection. SIB5 provides information for inter-system cell reselection toward 4G (LTE). SIB6 and SIB7 provide information for earthquake and tsunami warning systems (ETWS). SIB8 provides information for commercial mobile alert service (CMAS) notifications, for example, to deliver warning text messages to UE3. SIB9 includes information on harmonized universal time (UTC), global positioning system (GPS) time (for example, for GPS initial setup), and local time.

[0121] SIBs may be broadcast periodically (for example, according to a predetermined periodic pattern), or alternatively, they may be provided "on demand" upon request from, for example, UE3. For example, MIBs may be transmitted with a period of 80ms and repetitions occurring within 80ms, while SIB1 may be transmitted with a period of 160ms and a variable transmission repetition period (e.g., 20ms) within 160ms. SIB1 can be used to indicate to UE3 which SIBs are transmitted periodically and which SIBs are available on demand upon request from UE3. UE3 may be configured to request on-demand SIBs using message 1 (MSG1), which may be called an MSG1-based on-demand SI request, or message 3 (MSG3), which may be called an MSG3-based on-demand SI request.

[0122] A physical broadcast channel (PBCH) can be used to broadcast the MIB. Base station 5 may transmit the PBCH in an SS / PBCH block along with a synchronization signal (SS) (e.g., a primary synchronization signal (PSS) and a secondary synchronization signal (SSS)). An SS / PBCH block comprises four orthogonal frequency-division multiplexed (OFDM) symbols that map to the PSS, SSS, and PBCH associated with a demodulation reference signal (DM-RS). In the frequency domain, an SS / PBCH block contains 240 consecutive subcarriers. When UE3 is in an RRC connection state, base station 5 may provide UE3 with notification of the resources used for the SS / PBCH, for example, using dedicated signaling. SIB1 may be transmitted using a physical downlink shared channel (PDSCH). OSI may similarly be transmitted using a PDSCH, for example. If one or more beamformed transmissions are transmitted in a cell provided by base station 5, only some of the SIs (such as some SIBs) may be transmitted using specific beams or using specific transmission / reception points (TRPs).

[0123] (UE Mobility) Figure 5 shows an overview of a mobility procedure that can be performed in the type of communication system 1 illustrated in Figure 1. In this example, a handover of the UE3 from the source base station 5 to the target base station 5 is performed.

[0124] In the optional step S501, UE3 performs a measurement. The measurement may be a measurement of the signal transmitted by source(R)AN node 5, or a measurement of the signal transmitted by target(R)AN node 5. The measurement may be a measurement of signal intensity, which can be used as part of UE3's determination that a handover should be made from source(R)AN node 5 to target(R)AN node.

[0125] In the optional step S502, UE3 sends a measurement report to source(R)AN node 5 providing notification of the measurement results. The measurement report may also be sent from UE3 to source base station 5 in an RRC message. In this example, source(R)AN node uses the information provided in the measurement report to determine whether UE3 should be handed over to target(R)AN node 5. However, it will be understood that the determination that a handover to target(R)AN node should be performed may alternatively (or additionally) be based on measurements performed at source(R)AN node 5 or target(R)AN node 5. Alternatively, the determination that a handover of UE3 should be performed may be based on factors other than signal measurements, such as the level of density in the cell operated by source(R)AN node 5, or inferences (e.g., decisions or predictions) generated using an AI / ML model.

[0126] In step S503, Source(R)AN node 5 sends a handover request to Target(R)AN node 5, requesting a handover of UE3 from Source(R)AN node 5 to Target(R)AN node 5. The handover request may include, for example, notification of Source(R)AN node 5's identity, the cause of the handover, the target cell's identity, UE3 context information (such as the UE3's maximum bitrate or security capabilities), and UE history information. If the handover was triggered by a measurement report received by Source(R)AN node 5 in step S502, the cause could indicate, for example, that the handover is desirable for radio reasons. Alternatively, if the handover was triggered to reduce the load on Source(R)AN node 5, the cause could indicate that the handover is to reduce the load on the serving cell. The handover request message may also include notification of the AMF10-1 servicing the UE3.

[0127] In step S504, the target (R)AN node sends a notification of receipt of the handover request (which may be referred to as a “Handover Request Receipt Notification” message). The Handover Request Receipt Notification message includes notification of the handover configuration information for the handover to be forwarded to UE3. The Handover Request Receipt Notification message may also include configuration information that enables the source (R)AN node 5 to begin forwarding user plane data for UE3 to the target (R)AN node 5.

[0128] The transmissions in steps S503 and S504 may be performed via the Xn interface between the source (R)AN node 5 and the target (R)AN node 5 (therefore, the handover procedure in this example may be called an Xn-based handover procedure). Steps S501 to S504 may be referred to as the "handover preparation phase".

[0129] In step S505, the source (R)AN node sends handover configuration information to the UE3. The configuration information for the handover may be, for example, the RRC configuration sent in an RRC configuration message or an RRC reconfiguration message. In step S506, the UE3 applies the configuration received for the handover and sends a notification to the target (R)AN node 5 that the configuration for the handover is complete. The message sent in step S505 may be, for example, an RRC Reconfiguration Complete message. Steps S505 to S506 may be referred to as the "handover execution phase".

[0130] Following the handover execution phase, UE3 can operate to send uplink transmissions to target(R)AN node 5 (e.g., uplink data) and receive downlink transmissions from target(R)AN node 5 (e.g., downlink data).

[0131] It will be understood that the mobility methods and handover procedures for UE3 are not limited to the example shown in Figure 5. For example, UE3 may be configured to perform a conditional handover (CHO) in which UE3 determines whether or not to hand over UE3 to a candidate cell based on one or more execution conditions. It will also be understood that handovers can be performed where DU50 changes but CU60 remains the same (inter-DU intra-CU handover), where both DU50 and CU60 change (inter-DU inter-CU handover), or between two cells operating with the same DU50.

[0132] (Random access) Figure 6 shows a random access (RA) procedure that can be performed in the system of Figure 1. The RA procedure can be used, for example, for initial access by UE3 in RRC idle mode or for transitioning from RRC inactive mode to RRC connected mode. The RA procedure may also be used during the handover of UE3 from the source base station to the target base station (for example, as in the handover procedure described above with reference to Figure 5) for initial access to the target base station 5.

[0133] In step S601, UE3 transmits a random access preamble to base station 5. In this example, UE3 selects a random access preamble to transmit from a group of random access preambles shared with other UE3s. The transmission in step S601 may also be referred to as message 1 (MSG1) and is transmitted using PRACH.

[0134] In step S602, base station 5 transmits a random access response to UE3. The transmission in step S602 may be referred to as message 2 (MSG2). The random access response indicates the time and / or frequency resources (e.g., resource blocks and / or symbols) that UE3 will use to transmit a subsequent transmission to base station 5. The random access response may also include further information for UE3 to use for communication with base station 5, such as a timing advance (TA) value.

[0135] In step S603, UE3 transmits a transmission to base station 5 using the notified time and / or frequency resources. The transmission in step S603 may be referred to as message 3 (MSG3). The transmission in step S603 may be a layer 2 (L2) or layer 3 (L3) message. The transmission in step S603 may include, for example, an RRC setup request, an RRC resume request, an RRC re-establishment request, or an RRC reconfiguration completion message.

[0136] If two UE3s select and transmit the same random access preamble in step S601, and receive and decode MSG2 transmitted by base station 5 in step S602, the two UEs can transmit MSG3 using the same time and / or frequency resources. This situation may be called a “conflict” or “collision”. To resolve the conflict, in step S604, base station 5 sends a content resolution message to UE3. The transmission in step S604 may be referred to as message 4 (MSG4). MSG4 indicates to UE3 whether the base station successfully received and decoded the MSG3 transmitted by UE3 in step S603. If base station 5 decodes an MSG3 transmitted by another UE3 that is in conflict with UE3, or if interference occurs between the MSG3s transmitted by the two UE3s, the MSG3 transmitted in step S603 may not have been successfully received or decoded by base station 5. If MSG3 transmitted by UE3 has not been decoded by base station 5 (UE3 can determine this if it has not received MSG4 from base station 5), UE3 returns to step S601 of the method and transmits another MSG1 to base station 5 (for example, after selecting a different random access preamble).

[0137] The procedure shown in Figure 6 is an example of a conflict-based RA procedure in which UE3 selects a random access preamble from a group of preambles that are also available to other UE3s (thus a conflict can occur if two UE3s select the same random access preamble). Alternatively, base station 5 may send the random access preamble assignment to UE3 before UE3 sends MSG1 to base station 5, in which case the RA procedure is conflict-free (and does not require conflict resolution in step S604). The random access preamble assignment may be sent to UE3 using an RRC message or layer 1 (L1) signaling (e.g., using DCI carried by PDCCH). In the method shown in Figure 5, in step S505, the random access preamble assignment for communication with target base station 5 may be sent to UE3.

[0138] MSG1 and / or MSG3 may be used by UE3 to request on-demand SI from base station 5.

[0139] (Broadcast and multicast) Base station 5 may transmit a broadcast intended for reception by any UE3 within base station 5's cell, or it may transmit a transmission (such as point-to-point, PTP, or transmit) intended for reception by a specific UE3. Base station 5 may also transmit a transmission (such as point-to-multiple, PTM, or transmit) intended for reception by a specific group of UE3. A transmission intended for reception by a single UE3 may be called a unicast transmission, and a transmission intended for reception by a group of UE3 may be called a multicast transmission.

[0140] A multicast service may include a PTP leg between base station 5 and a single UE3, and a PTM leg between base station 5 and multiple UE3s. PTP and PTM transmissions are schematically shown in Figure 7. Although UE3s are shown separately in Figure 7, it will be understood that UE3s can receive both the PTP and PTM portions of a multicast. PTP may be described as a PTP "leg" or "part" of a multicast transmission. Similarly, PTM may be described as a PTM "leg" or "part" of a multicast transmission.

[0141] A PTM leg has an MBS radio bearer (MRB) with a corresponding MRB configuration. Each MRB may have an associated identifier (e.g., MRB identity) that can be used to identify the MRB. The MRB identity may be included in any appropriate transmission for MRB configuration. Multicast services may be interrupted (the process by which the MRB is released) or reactivated based on multicast data activity (or inactivity). The configuration of one or more MRBs may be provided to the UE3 and / or base station using any appropriate radio link control (RLC) configuration signaling (e.g., in an RLC bearer configuration message).

[0142] Base station 5 may provide multicast MRB configuration to UE3 via dedicated signaling. The multicast MRB may be configured in DL-only RLC unacknowledge mode (RLC-UM) where acknowledge / negative-acknowledge (ACK / NACK) feedback is not transmitted, or the MRB may have a bidirectional RLC-UM configuration for PTP transmission.

[0143] A multicast MRB configuration may include an RLC-acknowledge mode (RLC-AM) configuration for sending ACK / NACK feedback. A multicast MRB configuration may also include an RLC-unacknowledge mode (RLC-UM) configuration for not sending ACK / NACK feedback.

[0144] A multicast MRB configuration may include RLC-AM entities for PTP transmissions. A multicast MRB configuration may include DL-only RLC-UM entities for PTM transmissions.

[0145] A multicast MRB configuration may include two RLC-UM entities. One RLC-UM entity may be a DL-only RLC-UM entity for PTP transmission, and the other RLC-UM entity may be a DL-only RLC-UM entity for PTM transmission.

[0146] A multicast MRB configuration can include three RLC-UM entities, one of which is a DL-only RLC-UM entity, one of which is a UL RLC-UM entity for PTP transmission, and the other RLC-UM entity is a DL-only RLC-UM entity for PTM transmission.

[0147] A multicast MRB configuration may include two RLC entities, one of which is an RLC-AM entity for PTP transmission and the other is a DL-only RLC-UM entity for PTM transmission.

[0148] (Logical channels and logical channel priority) A logical channel (LCH) may be a control channel for transmitting control and / or configuration information (control plane information), or it may be a channel used for transmitting user data (user plane information). Logical channels that may be used in the system shown in Figure 1 include a broadcast control channel (BCCH), a paging control channel (PCCH), a common control channel (CCCH) used by the UE3 during initial access, a dedicated control channel (DCCH), and a dedicated traffic channel (DTCH). The system in Figure 1 may also use one or more transport channels. Transport channels include a broadcast channel (BCH), a paging channel (PCH), a downlink shared channel (DLSCH), an uplink shared channel (ULSCH), and a random access channel (RACH). Mapping between logical channels and transport channels may be performed at the medium access control (MAC) layer, and multiple logical channels may be multiplexed for transmissions using transport channels (for example, based on the priority of each logical channel, as described later). For example, BCCH may be mapped to BCH or DLSCH, and PCCH may be mapped to PCH. Transport channels are mapped to corresponding physical channels (e.g., PDCCH, PDSCH, or PBCH for downlink transmission; or PUSCH, PUCCH, or PUSCH for uplink transmission).

[0149] A logical channel can be identified using its corresponding logical channel ID (LCID). A set of logical channels can be grouped into a logical channel group (LCG), which can be identified using its corresponding index (e.g., an index between 0 and 7).

[0150] Logical channels can be prioritized by the network (for example, by transmit priority). For example, a logical channel used as part of a handover procedure may be assigned a relatively high transmit priority because a transmit delay in the handover procedure increases the likelihood of a handover failure. The base station 5 may decide to prioritize including data (or other information) corresponding to a higher-priority logical channel in the medium access control (MAC) protocol data unit (PDU) for transmission to the UE3, rather than including data or other information corresponding to a lower-priority logical channel. The base station 5 can also control the scheduling of uplink transmissions by the UE3 based on logical channel priority.

[0151] A prioritized bit rate (PBR) may be defined for a logical channel. The prioritized bit rate may be set by base station 5. The prioritized bit rate is the bit rate configured for use on higher-priority logical channels, and the remaining available bit rate (or a portion of the remaining available bit rate) is configured for transmission of lower-priority logical channels. The use of PBR is beneficial in avoiding situations where only the highest-priority logical channels are transmitted.

[0152] (Artificial Intelligence(AI) / Machine Learning(ML)) Figure 8 illustrates the framework for AI / ML models and how the various entities within the framework can interact with each other.

[0153] The entity includes a data acquisition function 41, a model training function 43, a model inference function 45, and an actor 47. The data acquisition function 41 provides input data (training data) to the model training function 43 and the model inference function 45. The data to be collected may be, for example, mobility data (such as UE3 handover or UE3 location). For example, the data may be acquired by base station 5 (for example by receiving measurement reports from UE3 or by receiving data from another base station 5 or core network node / function) and transmitted to another base station 5 that generates AI / ML model inference output (or alternatively, the same base station that acquired the data may generate the AI / ML model output). The model training function 43 can perform ML model training, validation, and testing and generate model performance metrics as part of the model testing procedure. The model inference function 45 provides an AI / ML model inference output (e.g., prediction or decision), and actor 47 is a function or node that receives the output from the model inference function 45 and triggers or performs a corresponding action (e.g., a base station 5 that increases / decreases its transmission power or initiates a handover procedure regarding UE3).

[0154] The AI / ML model inference output may be, for example, a prediction of the mobility of UE3 (such as expected path, route or trajectory, inter-cell or inter-beam mobility, or expected handover), or one or more parameters for use in encoding or decoding transmissions between base station 5 and UE3. The functions shown in Figure 8 may be located in the same place on a single node of the communication network (e.g., on base station 5 or a core network node / function), or they may be distributed among multiple network nodes (e.g., multiple base stations 5).

[0155] In the context of this framework, the terms referenced by 3GPP include the following: AI / ML training: An online or offline process for training AI / ML models. AI / ML validation: A method for evaluating the quality of an AI / ML model (e.g., prediction accuracy) using a different dataset than the one used to train the model. AI / ML model testing: A method for evaluating the performance of a final AI / ML model using a different dataset than the one used for training and validation. AI / ML Data Acquisition: Methods for collecting data by network nodes, management entities, and / or UE3 for training AI / ML models, for data analysis (e.g., model performance monitoring), and / or for generating inferences using AI / ML models. Model monitoring: A method for monitoring the inference performance (e.g., prediction accuracy) of AI / ML models. Training data: Data used as input for the AI / ML model training function. Supervised learning: A method for training AI / ML models using labeled data. Unsupervised learning: A method for training AI / ML models using unlabeled data. Semi-supervised learning: A method for training AI / ML models using both labeled and unlabeled data. Inference data: Data used as input for the AI / ML model inference function to generate inferences. Model Deployment / Update: Methods for deploying AI / ML models to model inference functions (e.g., sending them to network nodes), or for delivering updated models to model inference functions.

[0156] The data collection function 41 can be performed at various nodes in the communication network (for example, at one or more base stations 5 or UE3).

[0157] Figure 9 illustrates how to train an AI / ML model and monitor its performance. As shown in Figure 9, stored data / features may be extracted first in the data extraction step. In the data validation step, a decision is made (e.g., based on the extracted data) whether to proceed with training the AI / ML model or to hold off. In the data preparation phase, data is prepared for use in training the AI / ML model. For example, the data may be cleaned (e.g., filtered) or modified in any other appropriate way. The data may also be split into training data, validation data, and test data sets in the data preparation phase.

[0158] In the model training step, the AI / ML model is trained (or retrained) using the training data prepared in the data preparation step. It should be understood that any suitable training method can be used to train the AI / ML model (e.g., methods including supervised or unsupervised learning). In the model evaluation step, the AI / ML model is evaluated using a test dataset (which may be generated in the data preparation step) (e.g., the predictive accuracy of the AI / ML model is evaluated). In the model validation step, it is determined (e.g., based on the results of the model evaluation step) whether the AI / ML model is suitable for deployment on a communication network.

[0159] In the model service delivery step, the AI / ML model is deployed for use in communication system 1. AI / ML model deployment may include compiling the trained AI / ML model, packaging the model into an executable format, and delivering the AI / ML model to target devices. For example, the AI / ML model may be sent to base station 5 and / or UE3 for use by the base station and / or UE to generate predictions or decisions using the AI / ML model as part of the prediction service step, as shown in the figure.

[0160] In the performance monitoring step, the performance of the deployed AI / ML model is monitored. The predictive performance of the AI / ML model can be monitored by comparing the predictions generated using the model with one or more measurements. For example, if the AI / ML model is used to predict the location of UE3, the predictive accuracy of the AI / ML model can be evaluated using measurements of the actual location of UE3. Alternatively, if the AI / ML model is used to determine the parameters used for encoding and decoding data transmitted between base station 5 and UE3, the model may be evaluated based on the performance of the encoding and / or decoding process. In the retraining trigger step, retraining of the AI / ML model is triggered (for example, because the predictive accuracy of the AI / ML model falls below an acceptable threshold accuracy, or because the performance of the method using inference from the AI / ML model falls below an acceptable threshold performance), and the method returns to the data extraction step.

[0161] As described above with reference to Figure 8, each step of the method in Figure 9 may be performed on a single node of the communication system 1, or alternatively, the steps of the method may be distributed across multiple different nodes.

[0162] As described above with reference to Figures 8 and 9, information collected by nodes / functions within the communication network can be used as training data for AI / ML models and as inference data for generating one or more model inferences using the AI / ML models. The information used as training data and / or for generating one or more model inferences is sometimes referred to as "AI / ML information." Next, we will describe how to request and send AI / ML information.

[0163] Figure 10 shows an example of an AI / ML information request and AI / ML response. In step S1501, the first base station 5-1 transmits an AI / ML information request to the second base station 5-2. The AI / ML request is a request for AI / ML information from the second base station 5-2 (for example, information about the actual mobility of UE3).

[0164] After receiving the AI / ML information request in step S1501, the second base station 5-2 transmits an AI / ML information response containing AI / ML information to the first base station 5-1. Alternatively, the second base station 5-2 may initiate periodic reporting of AI / ML information to the first base station 5-1 in response to receiving the AI / ML information request. The periodic reporting may be configured using the corresponding AI / ML information reporting configuration indicated by the AI / ML information request (including, for example, the reporting frequency, the number of reports, or the reporting duration / period). The AI / ML information request may include an information element (IE) indicating that the second base station 5-2 is initiating or stopping periodic reporting of AI / ML information to the first base station 5-1. Alternatively, the AI / ML information request may be a request for a single report of AI / ML information from the second base station 5-2, rather than periodic reporting.

[0165] If the second base station 5-2 is unable to transmit the requested AI / ML information to the first base station 5-1 (for example, because the requested information is unavailable at the second base station 5-2), base station 5-2 may transmit a corresponding notification to the first base station 5-1 indicating that the second base station is unable to provide the requested information, such as an AI / ML information failure message. The AI / ML information failure message may include a notification of the reason (e.g., cause value) why the second base station 5-2 is unable to provide the requested AI / ML information.

[0166] Upon receiving AI / ML information, the first base station 5-1 may use the AI / ML information to train (or update) a corresponding AI / ML model (for example, with respect to UE3 mobility), or to generate predictions (for example, predictions of UE3 mobility). Alternatively, the first base station 5-1 may forward the AI / ML information to another network node for use with the AI / ML model at that network node.

[0167] In the example shown in FIG. 10, the AI / ML information response may include the requested AI / ML information. Instead, the AI / ML information response may be a notification that the second base station 5-2 will transmit the AI / ML information in a subsequent AI / ML information update (such as a receipt notification of the AI / ML information request). FIG. 11 is a diagram showing an example of AI / ML information update. In step S1601, the second base station 5-2 determines to transmit the AI / ML information update to the first base station 5. For example, the second base station 5-2 may determine to transmit the AI / ML information update to the first base station 5-2 based on the reporting period received by the second base station 5-2 in step S1501 of FIG. 10, or may determine to transmit the AI / ML information update based on a change in the AI / ML information stored in the second base station 5 (or based on the AI / ML information obtained at the second base station 5-2). In step S1602, the second base station 5-2 transmits the AI / ML information to the first base station 5-1 in the AI / ML information update.

[0168] <Distributed AI / ML Architecture> The network can include a primary node / function that hosts an AI / ML model and generates AI / ML model inferences. Alternatively, the AI / ML model may be distributed across various nodes within the network. For example, multiple base stations 5 may host an AI / ML model and generate inferences. This can increase the processing required at some network nodes, but when the AI / ML model is distributed among network nodes, the number of inferences transmitted between nodes decreases.

[0169] When an AI / ML model (or multiple AI / ML models - not necessarily the same model at each node) is provided to multiple base stations 5, feedback information can still be provided to each base station that uses the AI / ML model to generate inferences (e.g., to verify the accuracy of the model as described above).

[0170] <Configuration Information for AI / ML> Configuration information of an AI / ML model (sometimes referred to as "AI / ML configuration information") may be exchanged between nodes in the communication network. For example, a core network node can send AI / ML configuration information to base station 5, which hosts the AI / ML model.

[0171] AI / ML configuration information may include a list of supported use cases for AI / ML models (AI / ML models do not necessarily have to be for predicting UE mobility). Supported use cases include, for example, energy saving, traffic steering, anomaly detection, quality of experience (QoE) optimization, mobility robustness optimization (MRO), RAN slice service level agreement (SLA) assurance, large-scale multiple-input multiple-output (MIMO) beamforming optimization, network slice subnet instance (NSSI) resource allocation, optimized coverage and capacity optimization (CCO), mobility load balancing (MLB), RACH optimization, or UE transmit power optimization. AI / ML configuration information may include notifications of specific AI / ML models to use for specific use cases. AI / ML configuration information may also include notifications of whether feedback is needed (e.g., from another network node). Feedback may include, for example, communication performance feedback (e.g., notifying communication performance for communication between UE3 and base station 5).

[0172] If multiple AI / ML models are stored in UE3 (or base station 5, or other network nodes), UE3 can receive notifications about which AI / ML model to use. UE3 can also receive notifications (e.g., from base station 5) that a particular model should be activated or deactivated (for example, in response to a decision in the performance monitoring step in Figure 9, a particular AI / ML model can be deactivated if its prediction accuracy falls below an acceptable accuracy threshold). UE3 may have multiple AI / ML models, each intended for use in a specific scenario or configuration.

[0173] (Single-sided and double-sided models) The AI / ML model may be hosted (stored for generating inference) on both base station 5 and UE3, hosted only on base station 5, or hosted only on UE3. When the AI / ML model is used only on UE3, it may be referred to as a "one-sided" model. For example, UE3 may host an AI / ML model to generate time (e.g., time resources) for communication using a specific beam transmitted by base station 5. However, it should be understood that even if the model is a one-sided model, the model does not necessarily have to be trained on UE3. For example, the model may be trained on base station 5 or another node in the network (e.g., a core network node / function) and then sent to UE3 for use on UE3. In other words, the AI / ML model may be trained on another network node and then transferred / deployed to UE3.

[0174] Alternatively, the AI / ML model may be a "two-sided" model in which one AI / ML model is hosted on UE3 and the corresponding AI / ML model is hosted on base station 5 (however, the model does not necessarily have to be hosted on both UE3 and base station 5, and any two other suitable network nodes may be used instead). The AI / ML model hosted on UE3 and the AI / ML model hosted on base station 5 may be the same AI / ML model (however, they do not necessarily have to be the same model). UE3 can use the AI / ML model to generate a first inference, and base station 5 can use the AI / ML model to generate a corresponding second inference. For example, the first inference may be an inference of parameters used to encode or compress data (e.g., channel state information (CSI)) transmitted from UE3 to base station 5, and the second inference may be an inference of parameters used to decode or decompress the data at base station 5. As with the one-sided model, the two-sided model (or multiple models) can be trained on any suitable network node and then transmitted to UE3 and base station 5.

[0175] (AI / ML model acquisition) Next, we will explain how to deploy the AI / ML model. In this example, the AI / ML model is sent to UE3 for use in UE3. The AI / ML model may be a bifacial model (if the corresponding AI / ML model, or the same AI / ML model, is used in base station 5), or it may be an AI / ML model used only in UE3.

[0176] In this example, when UE3 is in an RRC idle or RRC inactive state, broadcast or multicast transmission is used to send the AI / ML model to UE3. When UE3 is in an RRC connected state, multicast transmission and / or RRC messages (such as dedicated RRC messages) are used to send the AI / ML model to UE3.

[0177] Figure 12 shows an example of how base station 5 broadcasts notifications for supported AI / ML models.

[0178] In step S1401, base station 5 broadcasts a notification of supported AI / ML models. In this example, the notification of supported AI / ML models is included in the system information (SI) broadcast in the base station's cell. In this example, UE3 is in an RRC idle or RRC inactive state (but could alternatively be in an RRC connected state). Therefore, advantageously, UE3 can receive information indicating which AI / ML models are supported by base station 5, even when UE3 is in an RRC idle or RRC inactive state.

[0179] The broadcast SI may include a list of AI / ML model IDs and / or version numbers of supported AI / ML models. Supported AI / ML models may be indicated on a per-use-case basis. For example, a first notification of supported AI / ML models for beam management may be provided, and a second notification of supported AI / ML models for encoding / decoding of the CSI may also be provided. The notification of supported AI / ML models may be broadcast periodically by the base station 5, or alternatively, on-demand in response to requests from the UE3. The broadcast SI may also include a notification of how the AI / ML models are obtained (e.g., signaling-based transmission between the UE3 and the RAN node 5, or database transmission between the UE3 and the AI / ML server 151). If the UE3 obtains the AI / ML models from the AI / ML server 151, the identity and (IP) address of the AI / ML server 151 may also be included in the SI. In this example, the notification in step S1401 is broadcast by the base station 5, but alternatively, the notification may be sent to the UE3 by multicast transmission.

[0180] In step S1402, UE3 determines whether to acquire one of the supported AI / ML models based on the notification of supported AI / ML models received from base station 5. In this example, UE3 determines to acquire one of the models and, in step S1403, sends a request for the model to base station 5. Step S1403 may be performed when UE3 is in an RRC idle or RRC inactive state (or, as part of a transition from an RRC idle or RRC inactive state to an RRC connected state, for example using MSG3, as will be described in more detail below). In step S1404, base station 5 sends the requested model to UE3. As will be described in more detail below, UE3 may be in an RRC connected, RRC inactive, or RRC idle state when it receives the AI / ML model from base station in step S1404.

[0181] In the example in Figure 12, UE3 sends a request for an AI / ML model to base station 5 and receives the requested AI / ML model from base station 5, but this is not necessarily required. Alternatively, after receiving notification of a supported AI / ML model, UE3 may request and receive an AI / ML model from any other suitable node in the network (e.g., another base station 5, or a core network node / function / server).

[0182] For example, Figure 13 shows a modification of Figure 12 in which UE3 requests an AI / ML model to be stored in the AI / ML server 151. Figure 13 includes new steps S1403b and S1403c. In step S1403b, base station 5 sends a request for the AI / ML model requested by UE3 to the AI / ML server 151. In step S1403c, AI / ML server 151 sends the requested model to base station 5 for transfer to UE3 in step S1404. The transfer of the AI / ML model through base station 5 in steps S1403c and S1404 may be transparent to base station 5 (for example, the AI / ML model can be transmitted using one or more transparent containers). In a further alternative, UE3 can retrieve the AI / ML model from the AI / ML server 151 via AMF10-1 using, for example, NAS-based signaling. For example, UE3 can send the AI / ML model request to AMF10-1 instead of sending it to base station 5. AMF10-1 can then request the model from AI / ML server 151, and AI / ML server 151 can transfer the AI / ML model to UE3. In another alternative, UE3 can request the AI / ML model from base station 5, and then base station 5 can request the AI / ML model from AI / ML server 151. However, instead of sending the AI / ML model to UE3 via the base station that received the request, the AI / ML model can be sent to UE3 via AMF10-1 (using corresponding NAS signaling). Instead of sending the AI / ML model request to base station 5, UE3 can alternatively send the AI / ML model request directly to the AI / ML server (for example, if UE3 has already obtained information indicating that the AI / ML model is stored in AI / ML server 151).

[0183] When UE3 requests an AI / ML model stored in AI / ML server 151, the transfer of the AI / ML model from AI / ML server 151 to UE3 may be a normal data transmission, and therefore, UE3 retrieving the AI / ML model from AI / ML server 151 may be transparent to the wireless network from a signaling perspective. However, if UE3 establishes an RRC connection with the wireless network for such data transmission, UE3 may include the RRC establishment cause and / or the AI / ML server address in the RRC message (for example, in the case of AI / ML model transfer). (R) RAN node 5 may transfer information to the core network. Usefully, this information helps RAN node 5 and / or core network nodes to establish a subsequent user plane data tunnel for AI / ML model transmission between AI / ML server 151 and UE3.

[0184] The determination of whether or not to retrieve an AI / ML model in step S1402 may be based on a comparison between the AI / ML models stored in UE3 and the supported AI / ML models. For example, base station 5 may provide notification of the model version of a supported AI / ML model in the information broadcast in step S1401, and UE3 may compare the version number of the model stored in UE3 with the version number of one of the supported models and determine that a newer version of the model is to be retrieved. Alternatively, UE3 may determine that UE3 does not store an AI / ML model for a particular use case (e.g., for encoding CSI) and therefore determine to retrieve a supported AI / ML model for that use case. Additionally or alternatively, UE3 may determine to send a request for an AI / ML model based on a timer. By using a timer, UE3 can request a newer version of the AI / ML model even if UE3 has not received notification of a supported AI / ML model in step S1401 (for example, UE3 may, based on the timer, send a request to base station 5 for the latest version of the AI / ML model stored in UE3, regardless of whether UE3 has received the transmission in step S1401). In a further alternative, base station 5 may, based on the timer, decide to send an updated version of the model to UE3 in step S1404. Thus, base station 5 can provide UE3 with a newer version of the AI / ML model even if base station 5 has not received a request for a newer version of the AI / ML model from UE3. This may be particularly useful for bifacial models where the version of the model in UE3 (for example, for encoding a CSI) may need to match or correspond to the version of the model in base station 5 (for example, for decoding a CSI).

[0185] By sending an AI / ML model request in step S1403, or by sending a model transfer based on a timer in step S1404, the risk of the UE3's model not matching the base station 5's model is reduced. Even when a timer is used for the transmission in S1403, it will be understood that the UE3 may still decide to send one or more AI / ML mode requests even if the time has not yet expired (for example, based on the information received in step S1401 as described above).

[0186] UE3 may request an AI / ML model from base station 5 by performing a random access procedure (for example, the RA procedure described above with reference to Figure 6). In this example, the MSG3 sent from UE3 to base station 5 in the RA procedure includes an RRC establishment cause indicating that UE3 is requesting an AI / ML model (for example, by including a notification of the identity of the requested AI / ML model, or a notification that UE3 should enter an RRC connection state in order to download the AI / ML model from base station 5). UE3 may request an AI / ML model using the RA procedure in both the example in Figure 12, where the requested AI / ML model is first stored in base station 5, and the method in Figure 13, where the AI / ML model is first stored in AI / ML server 151 (or any other suitable network node).

[0187] While the use of MSG3 and RRC establishment causes provides an efficient mechanism for indicating that UE3 is requesting an AI / ML model, alternatively, notification can be provided by any other suitable transmission from UE3 to base station 5. For example, UE3 can indicate that UE3 is requesting an AI / ML model using an RRC message (e.g., a dedicated RRC message). Any other suitable method for obtaining the AI / ML model can be used alternatively, and UE3 is not necessarily required to use the RA procedure to obtain the model.

[0188] In a further alternative, instead of UE3 requesting an AI / ML model (entering an RRC connection state to receive the model) in response to the determination in step S1402, UE3 may wait until it next enters an RRC connection state before acquiring an AI / ML model from base station 5. In another alternative, UE3 may receive an AI / ML model while it is in an RRC idle or RRC inactive state, rather than entering an RRC connection state to receive the AI / ML model. In this case, base station 5 sends a notification of the communication resources (e.g., time and frequency resources) that UE3 will use to receive the AI / ML model while UE3 is in an RRC idle or RRC inactive state.

[0189] The AI / ML model may be transmitted from base station 5 to UE3 in step S1404 using an RRC message or a user-plane transmission (e.g., using a data radio bearer (DRB)). Advantageously, a priority (e.g., transmission priority) can be assigned to the DRB or logical channel carrying the AI / ML model. As described above, a logical channel may be assigned an index indicating the transmission priority of the logical channel (e.g., by base station 5), and / or a prioritized bit rate (PBR). The priority or PBR configured for the DRB or LCH carrying the AI / ML model may depend, for example, on the type of AI / ML model required (e.g., the use case of the AI / ML). For example, a DRB or LCH used to transmit an AI / ML model for use as part of a handover procedure may be assigned a higher priority (or higher PBR) than when the AI / ML model is used in a beam prediction procedure.

[0190] From an air interface perspective, when AIML model transmission receives user plane transmission as described above, it may differ from conventional user plane (UP) transmission. Conventional UP transmission requires two or more part-based transmissions (air interface and backhaul-based fixed network), for example, a DRB over the air interface and, in addition, a data tunnel established between the base station 5 and the UPF in the core network that bridges the data to the data server. In this conventional method of UP transmission, base station 5 is not a data producer but rather a data "consumer" because the base station simply translates one or more QoS flows to the DRB at the SDAP layer to support data transmission for specific QoS services with respect to the data radio bearer over the air interface. However, this conventional UP transmission can be advantageously modified so that base station 5 can become a data producer if base station 5 itself holds the AIML model ready for transmission to the UE. When base station 5 determines to transfer the AI / ML model to UE3 via a UP-based channel, base station 5 can configure the data content of the AI / ML model as a Service Data Unit (SDU) to the PDCP layer, which can be viewed as a special data radio bearer. In this case, the AI / ML model data is not carried by the SDAP layer, unlike in conventional methods.

[0191] Step S1404 in Figures 12 and 13 may include transmitting the AI / ML model to the UE3 using a dedicated radio bearer (e.g., a bearer other than the old SRB / DRB). The AI / ML model can be transmitted using a logical channel (e.g., a dedicated logical channel). A priority and / or PBR can be assigned to the logical channel as described above, and alternatively, the logical channel does not have to be multiplexed with other logical channels and can instead transmit separately.

[0192] Figure 14 shows an example of how the requested AI / ML model can be retrieved by the UE3 when the requested AI / ML model is first stored in the distributed base station CU60. Steps S601 to S603 are the same as steps S1401 to S1403 described above and will not be repeated here. In step S604, the DU50 sends the request for the AI / ML model requested by the UE3 to the CU60, and in step S605, the CU60 sends the AI / ML model to the DU50. Step S606, in which the DU50 sends the requested AI / ML model to the UE3, is the same as step S1404 in Figures 14 and 15.

[0193] In step S605, an AI / ML model can be sent from CU60 to DU50 using a dedicated F1 application protocol (AP) message or procedure. Furthermore, CU60 can also send notifications of supported AI / ML models broadcast by DU50 in step S601 to DU50. Notifications of supported AI / ML models (e.g., model ID) can be sent from CU60 to DU50 using an F1-AP message (e.g., a dedicated F1-AP message). Thus, DU50 can determine the notification of a supported AI / ML model broadcast in step S601.

[0194] As described above, notifications of supported AI / ML models may be transmitted in step S1401 (or step S601) using system information broadcast in the cell of base station 5. An SIB can be used to transmit notifications of supported AI / ML models. This SIB may be referred to as the "AI / ML SIB". SIB1 can be used to provide notification that the AI / ML SIB is available for broadcast within the cell (the AI / ML SIB may also be an on-demand SI transmitted in response to a request from UE3, which is transmitted by UE3 before step S1401, although this is not shown in Figure 12). The MIB and SIB1 can provide notification of scheduling information for receiving and decoding dedicated AI / ML SIBs to UE3. The AI / ML SIB may contain the model ID of a supported (or "available") AI / ML model. As described above, the AI / ML SIB may indicate the supported AI / ML mods for each use case.

[0195] The AI / ML SIB may be broadcast periodically (for example, according to a predetermined periodic pattern), or alternatively, it may be provided "on demand" upon request from, for example, UE3. SIB1 can be used to indicate to UE3 whether the AI / ML SIB is being sent periodically or is available on demand.

[0196] If the AI / ML SIB is available on demand, base station 5 provides a notification of the AI / ML SIB's availability or information indicating the supported AI / ML model IDs for specific features (e.g., beam management) within the system information SIB1. UE3 may be configured to request the AI / ML SIB using message 1 (MSG1), which may be called an MSG1-based on-demand SI request for the AI / ML SIB, or message 3 (MSG3), which may be called an MSG3-based on-demand SI request for the AI / ML SIB. UE3 may also use another type of uplink message to indicate that UE3 is requesting information about one or more AI / ML models supported by the base station (e.g., AI / ML model IDs). When the network receives UE3's request for information (e.g., AI / ML IDs), the network broadcasts the supported AI / ML information (e.g., AI / ML model IDs) for one or more features requested by UE3 (e.g., using the system information block AI / ML SIB). Furthermore, UE3 can obtain AI / ML information by receiving and decoding broadcast messages (such as AI / ML SIBs).

[0197] Referring to Figures 12 to 14, in the examples described above, UE3 may request a single AI / ML model, or alternatively, multiple AI / ML models may be requested in step S1403 (or step S603).

[0198] (Area / location-based AI / ML models) Next, we will describe methods related to area-based or location-based AI / ML models. An AI / ML model may be intended for use in a specific area or location. An AI / ML model may be intended for use in a specific cell or group of cells that can be operated by one or more base stations 5. For example, an AI / ML model may be intended for use in a group of cells for beam management.

[0199] The areas in which an AI / ML model is used may include one or more cells, one or more RAN-based notification areas (RNAs), or registration areas (RAs). However, it should be understood that any other suitable areas for using an AI / ML model can be defined. Areas in which an AI / ML model is used for a specific function (e.g., beam management, CSI coding / decoding, or mobility) are sometimes called AI / ML model function areas.

[0200] The cells provided by base station 5 may be part of multiple AI / ML model functional areas. Figure 15 shows an example where a first base station 5-1 provides a first cell 180 and a second cell 181, and a second base station 5-2 provides a third cell 181. In this example, the first AI / ML model is intended for use in the first cell 180 and the second cell 181 for a first function (e.g., beam management). Therefore, the AI / ML model functional area of ​​the first AI / ML model includes the first cell 180 and the second cell 181. The second AI / ML model is intended for use in the second cell 181 and the third cell 182 for a second function (e.g., CSI coding / decoding). Therefore, the AI / ML model functional area of ​​the second AI / ML model includes the second cell 181 and the third cell 182.

[0201] In this example, the second cell 181 belongs to both the AI / ML model function area of ​​the first AI / ML model and the AI / ML model function area of ​​the second AI / ML model. In this example, one AI / ML model is used for each function in each area. Alternatively, two or more AI / ML models may be available for use in the functions of a particular area (for example, two or more AI / ML models may be available for UE mobility inference in a particular cell).

[0202] In this example, base station 5-1 is configured to send a broadcast transmission in the first cell 180 indicating that the first cell 180 belongs to the AI / ML model functional area of ​​the first AI / ML model, and to send a broadcast transmission in the second cell 181 indicating that the second cell belongs to both the AI / ML model functional area of ​​the first AI / ML model and the AI / ML model functional area of ​​the second AI / ML model. The notification of which AI / ML model functional area a cell belongs to may be called AI / ML model area information. Thus, a UE3 in a cell of base station 5-1 can determine which AI / ML model to use for a particular function within that cell.

[0203] Base station 5 may be configured to indicate the model function area to which a cell belongs, for each AI / ML model or for each function, in broadcast transmission. For example, base station 5 may support two AI / ML features / functions, where AI / ML model X is used for a first function and AI / ML model Y is used for a second function. From a network deployment perspective, AI / ML model X for the first function may belong to area N (which may be a relatively small area), and AI / ML model Y for the second feature may belong to area M (which may be a relatively large area). Broadcast information may indicate that the cell supports AI / ML models X and Y, that the cell supports a first function with AI / ML model X and a second function with AI / ML model Y, or alternatively, that the cell is part of corresponding areas N and M (for different models or functions).

[0204] Figure 16 shows how AI / ML model area information is received by UE3. In step S1901, base station 5 transmits (broadcasts) AI / ML model area information in its cell, and the information is received by UE3 in the cell.

[0205] In step S1902, UE3 determines to use a specific AI / ML model based on the AI / ML model area information. For example, if UE3 is in the second cell 181 in Figure 15 and receives AI / ML model area information indicating that a first AI / ML model is to be used for a first function of the second cell 181, UE3 determines to use the first AI / ML model for the first function of the second cell 181. If UE3 does not support the AI / ML model indicated in the AI / ML model area information, UE3 can simply ignore the AI / ML model area information. After UE3 has determined to use a specific AI / ML model, UE3 may retrieve the AI / ML model (if it is not already stored in UE3) by any of the methods described herein (for example, one of the methods shown in Figures 12 to 14). For example, as described above, UE3 may use a random access procedure including MSG3 as part of the method for retrieving the AI / ML model. As described above, UE3 may obtain AI / ML models directly from base station 5, or from another node in the network (for example, from AI / ML server 151 (via AMF10-1), from operations, administration, and maintenance server (OAM), or from any other suitable node / function in the network).

[0206] Alternatively, the AI / ML model area information may simply include a notification that a particular function is supported in that area, rather than AI / ML model area information that includes a notification of which AI / ML models are supported for a particular function. In this case, after receiving the AI / ML model area information, UE3 may decide to retrieve the system information broadcast in the cell to determine which AI / ML model to use. For example, as described above, UE3 may request an on-demand SIB that includes a notification of the AI / ML models supported for a particular function in the cell. UE3 may decide to retrieve the AI / ML model after moving to a new cell (e.g., following a cell reselection procedure) and receiving the broadcast transmission in step S1901, or if the AI / ML for use with a particular function in the cell has changed (this can also be identified by UE3 based on the broadcast transmission in step S1901). UE3 may be configured to periodically check for transmissions of AI / ML model area information by base station 5 (e.g., by receiving and decoding the corresponding SI) based on a timer. Similarly, base station 5 may be configured to periodically broadcast AI / ML model area information in one or more cells based on a timer.

[0207] If an AI / ML model supported for use in a specific area of ​​functionality is updated, UE3 can retrieve the updated model using one of the methods described above (for example, the method described with reference to Figure 14).

[0208] UE3 may store multiple AI / ML models that can be used for a particular function, and UE3 may select one of the multiple AI / ML models based on the AI / ML model area information received in step S1901. For example, UE3 may store a first AI / ML model for beam management in a first area and a second AI / ML model for beam management in a second area, and determine to use the first AI / ML model based on the notification in the AI / ML model area information that the cell belongs to the first area. If the cell belongs to both the first and second areas, UE3 can provide notification to the network (e.g., base station 5) whether to use the first or second AI / ML model (e.g., which model is preferred by UE3). Thus, advantageously, in the case of bifacial models, it is possible to avoid inconsistencies between the model used by base station 5 and the model used by UE3 when multiple models are supported for the same function within a particular area. UE3 can use a first RRC message transmitted to base station 5 after UE3 to provide notification of which AI / ML model to use (or which AI / ML is preferred to use). UE3 may include notification in MSG3 as described above, with reference to Figure 6.

[0209] In the example in Figure 15, UE3 may move from the second cell 181 to the third cell 182. Meanwhile, in the second cell 181, UE3 uses the first AI / ML model for the first function. However, in this example, the third cell 182 does not support the first function. Therefore, UE3 may decide not to use (or disable) the first AI / ML model for the first function after moving to the second cell 5-2. For example, UE3 may decide not to use (or disable) the first AI / ML model in response to receiving a broadcast transmission from the second base station 5-2 indicating the AI / ML models supported in the third cell 182 (or the AI / ML model function area to which the third cell 182 belongs). Alternatively, UE3 may decide not to use (or disable) the first AI / ML model in the third cell 182 even if UE3 has not received a broadcast transmission from the second base station 5-2. For example, UE3 can determine by default not to use (or disable) the first AI / ML model for the third cell 182, and only decide to use the first AI / ML model for the third cell if UE3 receives notification that the first AI / ML model is available for use on the third cell 182. Thus, advantageously, even if base station 5-2 is an older base station that may not support the transmission of AI / ML-related information, the use of unsupported AI / ML models or AI / ML model functions can be avoided.

[0210] (UE mobility and AI / ML models) As described above with reference to Figure 5, in handover procedure 5, UE3 may be handed over from source base station 5 to target base station. Next, an improved method for transmitting AI / ML models and / or AI / ML model-related information between base stations 5 and between base station 5 and UE3 will be described.

[0211] UE3 is initially connected to source base station 5-1 and configured to use an AI / ML model. For example, UE3 may use an AI / ML model for energy saving, traffic steering, anomaly detection, QoE optimization, or any other suitable AI / ML model. Alternatively, the AI / ML model may be a "two-sided" model where the AI / ML model is hosted on UE3 and the corresponding AI / ML model is hosted on base station 5. Alternatively, the AI / ML model may be a "one-sided" model used only on UE3. UE3 may acquire an AI / ML model using one of the methods described above, for example, referring to Figures 12 to 14.

[0212] After handover to target base station 5-2, UE3 can continue to use the same AI / ML model for specific features that UE3 was using before the handover. Alternatively, UE3 may switch to using a different AI / ML model for features after UE3 has been handed over to target base station 5-2. For example, source base station 5-1 and target base station 5-2 may be operated by different vendors, and the two base stations 5-1 and 5-2 may use different AI / ML models for specific features.

[0213] In a further alternative scenario, for example, if the target base station 5-2 does not support the use of AI / ML models, the use of AI / ML models for specific features or use cases may not be supported after UE3 is handed over to the target base station 5-2, in which case UE3 may not use AI / ML models for features at all.

[0214] Figure 17 shows a modified version of Figure 5, in which various steps of the method have been modified to include the transmission of an AI / ML model and / or AI / ML model-related information.

[0215] In step S1700, UE3 is connected to source base station 5-1. UE3 may transmit UL data to source base station 5-1 and receive downlink data from source base station 5-1.

[0216] Steps S1701 and S1702 are the same as steps S501 and S502 in Figure 5, and will not be repeated here. In this example, after receiving the measurement report from UE3 in step S1702, source base station 5-1 decides to hand over UE3 to target base station 5-2. However, it will be understood that source base station 5-1 may not necessarily receive the measurement report in step S1702 and may decide that UE3 should be handed over to target base station 5-2 by any other suitable method. For example, source base station 5-1 may decide to hand over UE3 to target base station 5-2 in order to reduce the RRC communication load at source base station 5-2, or based on the mobility of UE3 predicted using an AI / ML model (e.g., the predicted path of UE3).

[0217] In step S1703, source base station 5-1 sends a handover request to target base station 5-1. The handover request includes AI / ML information. The AI / ML information may include notification of the identity of the AI / ML models supported for use at the source base station. Notification of AI / ML model identity may be provided for each feature or use case that can be described by "functional identity". For example, AI / ML model information may provide notification that one or more AI / ML models are supported for UE mobility prediction, and that one or more AI / ML models are supported for anomaly detection (or for any other appropriate function or use case such as encoding / decoding CSI for sending / receiving CSI feedback reports, beam management methods, or UE position augmentation methods). The AI / ML information may include notification of the version numbers of the supported AI / ML models. It will be understood that the version numbers do not necessarily have to be the same as the AI / ML model ID numbers.

[0218] The AI / ML information transmitted in step S1703 may include notification of one or more AI / ML model functional areas to which the source base station cells that UE3 uses to communicate with source base station 5-1 before handover belong. The AI / ML model functional areas have been described above with reference to Figure 15.

[0219] The AI / ML information transmitted in step S1703 may include a notification of one or more AI / ML models currently selected (for example, by UE3 or source base station 5-1) for use by UE3 (or for use by source base station 5-1, or for use by both UE3 and source base station 5-1). Source base station 5-1 may support multiple AI / ML models for a particular feature, and source base station 5-1 may transmit a notification of the AI / ML models of the multiple AI / ML models currently in use for the feature.

[0220] The target base station 5-2 receives the AI / ML information transmitted in step S1703 and, advantageously, can determine one or more AI / ML models to be used after the UE3 is handed over to the target base station 5-2.

[0221] Although AI / ML information is shown in Figure 17 as being sent as part of the handover request, this is not necessarily required. For example, AI / ML information may be sent in separate transmissions between the source base station 5-1 and the target base station 5-2 between steps S1702 and S1703, or between steps S1703 and S1704. AI / ML information can be sent to the target base station 5-2 via the Xn interface between the source base station 5-1 and the target base station 5-2 using an Xn message, for example, a dedicated Xn message carrying the AIML information element (IE). Alternatively, AI / ML information may be sent to the target base station 5-2 using an RRC container in the handover request message in step S1703.

[0222] In step S1704, the target base station 5-2 sends a handover request acknowledgment to the source base station 5-1. The handover request acknowledgment is sometimes referred to as the “handover acknowledgment.” The handover request acknowledgment includes an AI / ML information response. The AI / ML information response may include a notification of one or more AI / ML models supported for use at the target base station 5-2. For example, the AI / ML information may include the AI / ML model ID or version number of the AI / ML models supported for use at the target base station 5-2 (for example, for each feature or use case that can be described by the “functional identity”).

[0223] For example, the AI / ML information transmitted in step S1703 may include a notification of one or more AI / ML models supported by source base station 5-1 for a particular function, and the AI / ML information response in step S1704 may include a notification of which of those AI / ML models are also supported by target base station 5-2. Thus, advantageously, the source base station can determine, based on the AI / ML information response, which AI / ML models are supported for UE3 by both source base station 5-1 and target base station 5-2 (which may be used by source base station 5-1 to improve the continuity of AI / ML model use during and after the handover procedure). The AI / ML information response may also include a notification of one or more additional AI / ML models supported by target base station 5-2 but not by source base station 5-1 (e.g., per feature or use case).

[0224] If the AI / ML information transmitted in step S1703 does not include a notification of one or more AI / ML models supported by source base station 5-1, or if no AI / ML information is transmitted from source base station 5-1 to target base station 5-2 in step S1703, target base station 5-2 may, in any case, transmit a notification of AI / ML models supported by target base station 5-2 in step S1704 (for example, per feature or use case).

[0225] Instead of sending a notification of a specific supported AI / ML model in step S1704 (for example, by sending a corresponding AI / ML model ID or version number), the target base station 5-2 may simply send a notification of a use case or feature of a supported AI / ML model in step S1704. For example, the AI / ML information response may include a notification that the target base station 5-2 supports the use of an AI / ML model for UE3 mobility prediction, or that the target base station 5-2 supports the use of an AI / ML model for anomaly detection (or any other appropriate use case / feature).

[0226] Similarly, the AI / ML information received from source base station 5-1 in step S1703 may include a notification of use cases or features of AI / ML models supported by source base station 5-1 (rather than a notification of specific supported AI / ML models). In this case, target base station 5-2 may, in step S1704, send a notification of AI / ML models supported by target base station 5-2 for one or more features / use cases indicated in the AI / ML information of step S1703.

[0227] The target base station 5-2 may also include in the AI / ML information response of step S1704 a notification of one or more AI / ML model functional areas to which the target base station's cell belongs. The AI / ML model functional areas may be indicated in the AI / ML information response for each AI / ML model or use case.

[0228] If different versions of the AI / ML model are supported by the source base station 5-1, the target base station 5-2 may include the version number of the AI / ML model supported by the target base station 5-2 in the AI / ML information response. For example, the AI / ML information received in step S1703 may include a notification that the first version of the AI / ML model is supported by the source base station 5-1. If the target base station 5-2 supports the second version of the AI / ML model instead of the first version, it may decide to send a notification to the source base station 5-1 that the target base station supports the second version of the AI / ML model (for example, by sending the version number of the second version of the model). If the target base station 5-2 supports the first version of the model, it may send a notification that the AI / ML model is supported (for example, using a 1-bit field) without necessarily sending a notification of the model's version number to the source base station 5-1. Alternatively, the target base station 5-2 may simply always send a notification of the supported AI / ML model's version number to the source base station in step S1704.

[0229] The AI / ML information response may include notification of one or more AI / ML models (or specific versions of AI / ML models) for use by UE3 during or after the handover. The one or more AI / ML models used by UE3 may be indicated per feature or use case. The AI / ML models indicated for use by UE3 during or after the handover may be the same models that UE3 was already using before the handover, or they may be different AI / ML models (for example, an AI / ML model supported by target base station 5-2 but not by source base station 5-1).

[0230] The AI / ML information response in step S1704 is shown in Figure 17 as being sent as part of the handover request acknowledgment, but this is not necessarily required. For example, the AI / ML information response may be sent in a separate transmission from the target base station 5-2 to the source base station 5-1 between steps S1703 and S1704, or between steps S1704 and S1705. The AI / ML information response may be sent to the source base station 5-1 via the Xn interface between the target base station 5-2 and the source base station 5-1 using an Xn message, for example, a dedicated Xn message containing the AIML information element (IE). Alternatively, the AI / ML information response may be sent to the source base station 5-1 using an RRC container in the handover request acknowledgment message in step S1704.

[0231] Based on the AI / ML information response received from the target base station 5-2 in step S1704, the source base station 5-1 can favorably determine one or more AI / ML models (or one or more versions of AI / ML models) for the UE3 to use after the handover of the UE3 from the source base station 5-1 to the target base station 5-2. For example, if the AI / ML information response includes notification of one or more models supported by the target base station 5-2 for a particular feature, the source base station 5-1 can compare the AI / ML models available at the source base station 5-1 (or UE3) with the AI / ML models supported by the target base station 5-2, and the UE3 can determine that it should use an AI / ML model that is available at the source base station 5-1 (or UE3) and supported by the target base station 5-2. Alternatively, source base station 5-1 (or target base station 5-2) may determine that UE3 will use an AI / ML model supported by target base station 5-2 but not currently available to source base station 5-1 or UE3 (or a version of an AI / ML model supported by target base station 5-2 but not available to source base station 5-1 or UE3). In this case, the AI / ML model to be used after the handover to target base station 5-2 may be transmitted to UE3 by source base station 5-1, target base station 5-2, or any other entity in the network.

[0232] In the optional step S1705, the target base station 5-2 transmits AI / ML model information to the source base station 5-1. The AI / ML model information includes information for obtaining an AI / ML model supported by the target base station 5-2. The AI / ML model information may include the AI / ML model itself, or it may provide a notification of how the AI / ML model can be obtained (for example, by providing the network address of a network entity or network server that can obtain the AI / ML model, such as an over-the-top (OTT) server or core network node / function that stores the AI / ML model).

[0233] If, in step S1705, the AI / ML model is transmitted from the target base station 5-2 to the source base station 5-1, then in the optional step S1706, the source base station 5-1 may transmit the AI / ML model to the UE3. The AI / ML model may be transmitted from the source base station 5-1 to the UE3 using any appropriate transmit. For example, the AI / ML model may be transmitted from the source base station 5-1 to the UE3 using an RRC message or user plane transmit (e.g., using a data radio bearer (DRB)). The AI / ML model may be transmitted to the UE3 using a dedicated radio bearer (e.g., a bearer other than an older SRB / DRB). The AI / ML model can be transmitted using a logical channel (e.g., a dedicated logical channel). Logical channels may be assigned priority and / or PBR as described above, and alternatively, logical channels do not necessarily have to be multiplexed with other logical channels but can be transmitted separately instead. Source base station 5-1 may encapsulate the AI / ML model for transmission to UE3 in the configuration for the handover message in step S1707 of Figure 17, which may be an RRC reconfiguration message, in which case it is not necessarily required to perform the other transmission in step S1706.

[0234] If the AI / ML model information provided in either step S1704 or step S1705 includes information for obtaining the AI / ML model (e.g., from an entity in the network other than the target base station 5-2) rather than the AI / ML model itself, the source base station 5-1 may obtain the AI / ML model before sending it to the UE3 (e.g., by requesting the model from another entity in the network, such as the AI / ML server 151, as shown in Figure 13). The transfer of the AI / ML model from the AI / ML server 151 to the UE3 via the source base station 5-1 may be transparent to the source base station 5-1 (e.g., the AI / ML model can be transmitted using one or more transparent containers). Alternatively, the source base station 5-1 may transfer information for obtaining the AI / ML model to the UE3, and the UE3 may obtain the model using the information for obtaining the AI / ML model (e.g., by requesting the model from the target base station 5-2 or from another entity in the network before, during, or after the handover procedure).

[0235] Instead of sending information for acquiring AI / ML models in step S1705, the information for acquiring AI / ML models can also be sent alternatively in the handover request acknowledgment message in step S1704. The transmission in step S1704 may include an explicit or implicit request for source base station 5-1 to send one or more AI / ML models to UE3. For example, the AI / ML information response in step S1704 may include a list of AI / ML models to be sent to UE3 (e.g., AI / ML model ID number, version number, and / or functional identity). The handover request acknowledgment message may include a notification of AI / ML models to be sent to UE3 for each feature or use case. Then, in step S1706, source base station 5-1 sends one or more AI / ML models to UE3 (without necessarily performing step S1705).

[0236] If an AI / ML model for use by UE3 after handover to target base station 5-2 is available to UE3 before or during the handover process, UE3 may begin using the AI / ML model before the handover is complete. For example, if UE3 is already using the AI / ML model before the handover (or the AI / ML model has been activated for use by UE3), source base station 5-1 may send a notification to UE3 that UE3 should continue using the AI / ML model after the handover is complete (or the AI / ML model should remain activated) (the notification may be provided, for example, in the transmission of step S1706 or S1707). The notification that UE3 will continue using the AI / ML model may include an AI / ML model ID number or version number for use with a particular feature (identified, for example, by a specific feature ID), or alternatively, a notification of a feature or use case (identified, for example, by a specific feature ID) for which UE3 will continue to use the same AI / ML model (identified, for example, by a specific feature ID) after the handover (and possibly during the handover).

[0237] The configuration for the handover transmitted in step S1707 may be included in the RRC reconfiguration message. The RRC reconfiguration message may include AI / ML model information (e.g., AI / ML model ID or version number) about the use cases or features of the AI / ML models supported by the target base station 5-2, based on the information received in step S1704 or S1705. If the target base station 5-2 includes a notification of preferred models for use by the UE3 after the handover to the target base station 5-2 in the transmission of step S1704 or S1705, the source base station 5-1 may transmit the notification of preferred models to the UE3 in step S1707 (or alternatively, the notification may be transmitted in step S1706). The notification of preferred models may be provided for each AI / ML use case or feature / feature. If, in step S1704 or S1705, the target base station 5-2 sends a notification to the source base station 5-1 that the target base station does not support a particular AI / ML model (this may be an explicit notification or an implicit notification, such as an AI / ML model not included in the list of models sent to the source base station 5-1), the source base station 5-1 may send a notification to the UE3 (in step S1706 or S1707) that the model not supported by the target base station 5-2 should not be used by the UE3 after the handover. The source base station 5-1 may also send a notification in step S1706 or S1707 to disable or deactivate the AI / ML model not supported by the target base station 5-2 in the UE3.

[0238] Following step S1709, UE3 performs an AI / ML update. For example, UE3 may deactivate or activate an AI / ML model based on information received from source base station 5-1 in step S1706 or S1707. If UE3 is already using an AI / ML model supported by target base station 5-2 for a particular feature, UE3 can simply continue using that AI / ML model. If UE3 receives an AI / ML model from source base station in step S1706, step S1709 may include preparing the model for use. For example, UE3 may configure one or more parameters of the received AI / ML model so that the AI / ML model is ready for use after (or possibly during) UE3's handover to target base station 5-2.

[0239] In step S1710, UE3 sends a notification that the configuration for handover to target base station 5-2 is complete. The transmission in step S1710 may be, for example, an RRC Reconfiguration Complete message. In the transmission in step S1711, UE3 may include a notification of a preferred AI / ML model (or preferred AI / ML model version) for a particular feature. For example, both UE3 and target base station 5-2 may support multiple AI / ML models (or AI / ML model versions) for a particular feature, and UE3 may indicate a preferred model or model version (for example, based on the memory or processing resources available in UE3, or based on the predictive accuracy of the model). If UE3 indicates a preferred AI / ML model, or if multiple AI / ML models or model versions are supported by UE3 and target base station 5-2 for other reasons, the target base station may send a notification of the AI / ML model (or model version) to be used in step S1711 (or in any other appropriate transmission) to UE3.

[0240] If the AI / ML model to be used by UE3 after handover is not received from source base station 5-1 and is therefore unavailable to UE3, UE3 may receive the AI / ML model from target base station 5-2 in step S1711.

[0241] In step S1712, the handover of UE3 from source base station 5-1 to target base station 5-2 is complete. UE3 can transmit uplink data to target base station 5-2, and target base station 5-3 can receive downlink data via UE3.

[0242] Because a relatively large amount of data is transmitted when sending an AI / ML model to UE3, the time required to transmit the AI / ML model to UE3 is relatively long. Therefore, when transmitting an AI / ML model from source base station 5-1 to UE3, there is a high risk that the handover of UE3 to target base station 5-2 may be completed before the transmission of the AI / ML model from source base station 5-1 to UE3 is complete.

[0243] If the transmission of the AI / ML model to UE3 is interrupted by a handover from UE3's source base station 5-1 to target base station 5-2, UE3 may be configured to discard the AI / ML model segment (or other received data unit) received from source base station 5-1, and transmission of the model (from target base station 5-2 to UE3) will resume after the handover to target base station 5-2 is complete.

[0244] Alternatively, the source base station may notify UE3 (for example, in step S1707 in Figure 17) that it will resume transmitting the AI / ML model following the handover. Following or during the handover, UE3 may provide target base station 5-2 with notification of the status of AI / ML model transmission (for example, by providing notification in the handover configuration complete message in step S1710 in Figure 17). For example, in step S1710, UE3 may send to target base station 5-2 notification of the number of AI / ML model segments (or any other appropriate data unit) received by UE3, notification of the last segment received by UE3, or any other appropriate information.

[0245] In the example in Figure 17, source base station 5-1, in any step S1708, stores a notification of the number of segments (or other appropriate data units) of the AI / ML model to be sent to UE3, or a notification of the last segment to be sent to UE3 (e.g., in RRC context information), and transmits the information to target base station 5-2 (e.g., via the Xn interface). Source base station 5-1 may also transmit information to target base station 5-2 indicating the identity of the AI / ML model that was transmitted. Source base station 5-1 may also transmit a notification to target base station 5-2 of the remaining parts of the AI / ML model to be sent to UE3, or the remaining parts (e.g., the segment number of the last segment of the AI / ML model transmission that was successfully transmitted to UE3, or the segment number of the next segment to be transmitted to UE3). Target base station 5-2 may then transmit the remaining parts of the AI / ML model to UE3 (e.g., in the optional step S1711 in Figure 17). Therefore, advantageously, the rest of the AI / ML model can be transmitted from the target base station 5-2 to the UE3, rather than the entire AI / ML model, reducing the amount of data that needs to be transmitted from the target base station 5-2 to the UE3.

[0246] Alternatively, instead of the source base station 5-1 sending a notification in step S1708 of the remaining portion of the AI / ML model to be sent to the UE3 by the target base station 5-2, the source base station 5-1 may send a notification to the target base station 5-2 indicating that the target base station will send the complete AI / ML model to the UE3 (or the target base station 5-1 may be configured to send the complete AI / ML model to the UE3 regardless of whether it receives a notification from the source base station 5-1). This may result in some duplication of data in the AI / ML model received by the UE3, but the AI / ML model can be sent to the UE3 more reliably. The source base station 5-1 may include the AI / ML model ID or version number in the transmission in step S1708. If the target base station 5-2 sends the complete AI / ML model instead of just the remaining portion of the model, the UE3 deletes the portion of the AI / ML model received from the source base station 5-1 (e.g., it is no longer remembered or cannot be overwritten). UE3 may delete the portion of the AI / ML model received from source base station 5-1 if it receives a notification from source base station 5-1 (for example, in step S1707) or from target base station 5-2 (for example, in step S1711) that the portion of the AI / ML model should be deleted.

[0247] The AI / ML model may be transmitted to UE3 via RRC transmission or user plane (UP) transmission. If the AI / ML model is transmitted to UE3 using one or more RRC messages (for example, if the message in step 1707 is an RRC reconfiguration message), the segments may be RRC segments, each having an associated RRC segment number. The RRC segment number of the last segment received by UE3 may be stored by UE3. The RRC segment number of the last segment to be transmitted to UE3 may be stored at source base station 5-1. If UE3 is receiving the AI / ML model from source base station 5-1 via user plane transmission, UE3 may receive the AI / ML model via a DRB established between source base station 5-1 and UE3. In this case, the segments may be PDCP segments with corresponding PDCP sequence numbers (SN), or they may be generated using a dedicated protocol layer (for example, the AI / ML protocol layer described above).

[0248] In step S1710, UE3 sends a notification to target base station 5-2 that UE3 has completed its configuration for handover to target base station 5-2. The handover configuration complete message may also be an RRC Reconfiguration Complete message.

[0249] In the example in Figure 17, the target base station 5-2 may, but is not required, send the AI / ML model to be used by UE3 after the handover to source base station 5-1. Alternatively, the target base station 5-2 may send the AI / ML model to UE3 in step S1711 after the handover to target base station 5-2 is complete. Advantageously, this may result in a delay before UE3 can begin using the AI / ML model (due to the time it takes for target base station 5-2 to send the AI / ML model to UE3), but the interruption of AI / ML model transmission by the handover procedure is beneficially avoided.

[0250] Please understand that UE3 handovers can occur from source base station 5-1, which does not support the use of AI / ML models with UE3, to target base station 5-2, which does support the use of AI / ML models with UE3; from source base station 5-1, which supports the use of AI / ML models with UE3, to target base station 5-2, which does not support the use of AI / ML models with UE3; or from source base station 5-1, which supports the use of AI / ML models with UE3, to target base station 5-2, which similarly supports the use of AI / ML models with UE3 (although not necessarily the same model, as described in Figure 17).

[0251] AI / ML models stored in UE3 may be part of the UE context. UE3 may have AI / ML models stored in its memory when UE3 is connected to source base station 5-1, and UE3 may be configured to retain the model in its memory even if the AI / ML model is not supported for use at target base station 5-2. This is beneficial, for example, to avoid the need for UE3 to reacquire the model if a further handover occurs to another base station that supports the AI / ML model (or back to source base station 5-1). UE3 may determine, based on its available memory, whether to retain an AI / ML model not supported at target base station 5-2 in its memory (e.g., one or more buffers). For example, if UE3's memory has sufficient space available to store an AI / ML model not supported at target base station 5-2, UE3 may decide to retain the model in its memory.

[0252] If UE3 uses an AI / ML model that is supported for use at target base station 5-2 following the handover, but the AI / ML model is unavailable at target base station 5-2 (e.g., stored), UE3 may transmit the AI / ML model to target base station 5-2 following or as part of the handover procedure (with any appropriate transmission from UE3 to target base station 5-2). This scenario may occur, for example, if target base station 5-2 includes notification in step S1704 of an AI / ML model that is supported for use at target base station 5-2 but is not currently stored at target base station 5-2. Alternatively, source base station 5-1 may transmit the AI / ML model to the target base station (or transmit information to the target base station for obtaining the AI / ML model, e.g., the network address used by target base station 5-2 to obtain the model). For example, if the AI / ML information in step S1704 indicates that the target base station 5-2 supports the AI / ML model but that model is not available at the target base station 5-2, the source base station may decide to transmit the model to the target base station 5-2 (in which case, instead of the source base station 5-1 receiving the model from the target base station 5-2 in step S1705, the source base station 5-2 transmits the model to the target base station 5-2 instead).

[0253] Even if the same AI / ML model for a particular use case or function is supported by both the source base station 5-1 and the target base station 5-2, the target base station 5-2 may support different versions of the model, or support the use of the model with a different set of parameters than the source base station 5-1. If the target base station 5-2 supports different versions of the model, in step S1704 or S1705, the target base station 5-2 may send a notification to the source base station (e.g., version number) of the supported version of the model. Similarly, a notification of the AI / ML model parameters supported by the target base station 5-2 may be provided in the transmission in step S1704 or S1705. If the handover request in step S1703 includes parameters used by the source base station 5-1 for a particular model, the target base station 5-2 may provide a notification of a subset of the parameters that will be changed for the use of the AI / ML model by the UE3 after the handover.

[0254] After receiving the AI / ML model version or AI / ML model parameters from the target base station 5-2, the source base station 5-2 may send a notification to the UE3 of the model version of the AI / ML model parameters to be used for a specific model after the handover (for example, in step S1706 or in the RRC reconfiguration message in S1707). Then, in step S1709, the UE3 can appropriately configure the AI / ML model based on the AI / ML version to be used or based on the AI / ML model parameters. If the AI / ML model to be used by the UE3 after the handover is already stored in the UE3 before the handover occurs, and only the AI / ML model parameters used with the AI / ML model are changed / updated, then the model is already stored in the UE3 and only needs to be reconfigured, so the UE3 does not necessarily need to receive the model from the source base station 5-1 or the target base station 5-2 in steps S1706, S1707, or S1711.

[0255] In the example shown in Figure 17, information indicating the AI / ML models supported by source base station 5-1 is transmitted in the handover request step S1703, and information indicating the AI / ML models supported by target base station 5-2 is transmitted in the handover request acknowledgment message in step S1704. However, it is not always necessary to transmit information indicating the supported AI / ML models as part of the handover procedure. Alternatively, for example, information indicating the supported models can be exchanged in an Xn setup procedure (for example, to initialize the Xn connection between the two base stations 5) or an Xn update procedure. The Xn setup procedure or Xn update procedure can be performed before the handover procedure in Figure 17, in which case the AI / ML information and AI / ML information response do not necessarily need to be transmitted in steps S1703 and S1704 (however, if the supported AI / ML models change after the Xn setup / update procedure, this can still be done to exchange more up-to-date AI / ML information, so the information can still be included).

[0256] Dedicated information elements may be included in Xn setup request messages, Xn setup response messages, or Xn update messages sent from source base station 5-1 to target base station 5-2, or from target base station 5-2 to source base station 5-1, to indicate one or more AI / ML models supported by base station 5 that sends messages (for example, for each use case or function, or for a particular UE3 or type / class of UE3). Dedicated information elements may include one or more bits to indicate whether a particular AI / ML function or capability is supported. Dedicated information elements may include one or more bits to indicate whether an AI / ML model is supported for a particular use case (for example, CSI feedback coding / decoding, beam management, or UE positioning accuracy method). Dedicated information elements may include one or more bits to indicate the model ID or version number of one or more supported AI / ML models. Dedicated information elements may include one or more bits to indicate that a base station cell is part of a particular AI / ML model function area. The dedicated information element may include one or more bits to indicate the model transfer method supported by the base station (for example, a model transfer architecture such as CP-based AI / ML model transfer or UP-based AI / ML model transfer).

[0257] If the information for obtaining the AI / ML model transmitted in step S1705 or S1706 includes information for obtaining the model from other nodes in the network (such as an AI / ML server, etc.), the transmission of the AI / ML model from the AI / ML server 151 to the UE3 may be performed as described above with reference to FIG. 13. The base station 5 in FIG. 13 may be either the source base station 5-1 or the target base station 5-2 in FIG. 17. The transmission of the AI / ML model from the AI / ML server 151 to the UE3 may be a UP-based transmission. When the transmission is a UP-based transmission, as shown in FIG. 13, there may be communication between the base station 5 and the AI / ML server 151 (such as a request for the AI / ML model transmitted from the base station to the AI / ML server 151), or there may be direct communication between the UE3 and the AI / ML server 151 (for example, the UE3 may directly transmit a request for the AI / ML model to the AI / ML server 151).

[0258] In step S1703 of FIG. 17, after the target base station 5-2 receives the handover request, the target base station 5-2 may send a notification to the AI / ML server 151 as to whether it is necessary to transmit the AI / ML model to the UE3 (or transmit a new version of the model, or updated parameters for use with the AI / ML model). Then, the AI / ML server 15 may transmit the AI / ML model to the UE3 via the source cell or the target cell (for example, as described with reference to steps S1403c and S1404 in FIG. 13, when the model is transmitted via the source cell, the base station 5 is the source base station 5-1, and when the model is transmitted via the target cell, the base station 5 is the target base station 5-2). Alternatively, for example, the AI / ML model can be transmitted from the AI / ML server 151 to the UE3 via another entity in the network, such as via the AMF10-1, using the corresponding NAS signaling as described above.

[0259] (User Equipment) FIG. 18 is a schematic block diagram showing the main components of the UE3 for the telecommunication system of FIG. 1.

[0260] As shown, the UE3 has a transceiver circuit 310 operable to transmit signals to and receive signals from the base station 5 via one or more antennas 330 (e.g., comprising one or more antenna elements). The UE3 has a controller 370 that controls the operation of the UE3. The controller 370 is associated with a memory 390 and is coupled to the transceiver circuit 310. Although not necessarily required for its operation, the UE3 may of course have all the normal functions of a conventional UE3 (such as a user interface 350, e.g., a touch screen / keypad / microphone / speaker, to enable direct control by the user and interaction with the user), which may be provided by any one or any combination of hardware, software, and firmware, as appropriate. The software may be pre-installed in the memory 390 and / or downloaded, e.g., via a communication network or from a removable data storage device (RMD).

[0261] In this example, the controller 370 is configured to control the overall operation of the UE3 by program instructions or software instructions stored in the memory 390. As shown, these software instructions include, among other things, an operating system 410, a communication control module 430, and an AI / ML module 450.

[0262] The communication control module 430 is operable to control communication between the UE 3 and one or more serving base stations 5 (and other communication devices connected to the base stations 5, such as additional UEs and / or core network nodes). The communication control module 430 is configured to handle uplink communication in general over associated uplink channels (e.g., via the physical uplink control channel (PUCCH), random access channel (RACH), and / or physical uplink shared channel (PUSCH)), including both dynamic and quasi-static signaling (e.g., such as SRS). The communication control module 430 is also configured to handle downlink communication in general over associated downlink channels (e.g., via the physical downlink control channel (PDCCH) and / or physical downlink shared channel (PDSCH)), including both dynamic and quasi-static signaling (e.g., such as CSI-RS). The communication control module 430 is responsible for, for example, determining where to monitor downlink control information (such as the locations of CSS / USS, CORESET, and associated PDCCH candidates to be monitored); determining resources to be used by UE3 for transmitting / receiving UL / DL communications (including interleaved resources and resources subject to frequency hopping); managing frequency hopping on the UE side; determining how slots / symbols are configured (e.g., for UL, DL, or SBFD communications); determining which one or more bandwidth portions are configured for UE3; determining how uplink transmissions should be encoded; and appropriately applying any SBFD-specific communication configurations. The communication control module 430 may be configured to control communications in any of the ways described above (for example, to transmit measurement reports in any of the ways described above).

[0263] The AI / ML module 450 can be configured to control the use of AI / ML models in UE3 (for example, to generate one or more inferences using the model). The AI / ML module 450 may be configured to perform any of the AI / ML-related functions in UE3 in any of the ways described above.

[0264] (base station) Figure 19 is a schematic block diagram showing the main components of base station 5 for communication system 1 shown in Figure 1. As shown, base station 5 has transceiver circuitry 510 for transmitting signals to and receiving signals from communication devices (such as UE3) via one or more antennas 530 (such as single or multi-panel antenna arrays / large antennas), and core network interface 550 (with, for example, N2, N3, and other reference points / interfaces) for transmitting signals to and receiving signals from network nodes in core network 7. Not shown, base station 5 may also be connected to other base stations via appropriate interfaces (such as the so-called "Xn" interface in NR). Base station 5 has a controller 570 that controls the operation of base station 5. Controller 570 is associated with memory 590. Software may be pre-installed in memory 590 and / or downloaded, for example, via communication system 1 or from a removable data storage device (RMD). In this example, the controller 570 is configured to control the overall operation of the base station 5 by program instructions or software instructions stored in the memory 590.

[0265] As shown in the figure, these software instructions include, among other things, the operating system 610, the communication control module 630, and the AI / ML module 650.

[0266] The communication control module 630 is operable to control communication between the base station 5, the UE3, and other network entities connected to the base station 5. The communication control module 630 is configured to generally control the reception and decoding of uplink communications over the relevant uplink channels (e.g., over the physical uplink control channel (PUCCH), random-access channel (RACH), and / or physical uplink shared channel (PUSCH)), including both dynamic and quasi-static signaling (e.g., SRS). The communication control module 630 is also configured to generally handle the transmission of downlink communications over the relevant downlink channels (e.g., over the physical downlink control channel (PDCCH) and / or physical downlink shared channel (PDSCH)), including both dynamic and quasi-static signaling (e.g., CSI-RS).

[0267] The communication control module 630 is responsible for managing full-duplex communications (such as SBFD), including, where appropriate, the separation of UL and DL communications over different physical antenna elements. The communication control module 630 is responsible for, for example, determining where UE3 should be configured to monitor downlink control information (such as the locations of CSS / USS, CORESET, and associated PDCCH candidates to be monitored); determining resources to be scheduled for UE transmission / reception of UL / DL communications (including interleaved resources and resources subject to frequency hopping); managing frequency hopping on the base station side; configuring slots / symbols appropriately (for example, for UL, DL, or SBFD communications); configuring one or more bandwidth portions for UE3; and providing relevant configuration signaling to UE3. The communication control module 630 may be configured to control communications in any of the methods described above (for example, to receive or transmit mobility information or handover requests for UE3).

[0268] The AI / ML module 650 may be configured to perform any of the AI / ML-related functions of UE3 in any of the methods described above. The base station 5 may be configured to train or retrain the AI / ML model as described above (for example, in response to UE mobility information fed back to base station 5 from another node in the network, such as another base station 5).

[0269] (Core network node / function) Figure 20 is a block diagram showing the main components of a core network node or function such as AMF, CPF, UPF, SMF, or OAM. As shown in the figure, the core network function includes a transceiver circuit 710 that can operate to transmit signals to and receive signals from other nodes (including UE3, base station 5, and other core network nodes) via the network interface 720. The controller 730 controls the operation of the core network function according to software stored in memory 740. The software may be pre-installed in memory 740 and / or downloaded, for example, via communication system 1 or from a removable data storage device (RMD). The software includes, among other things, an operating system 750 and a communication control module 760.

[0270] The communication control module 760 is responsible for handling (generating / transmitting / receiving) signaling between the core network functions and other nodes such as UE3, base station 5, and other core network nodes. The signaling may include, for example, UE3's UE context / UE capability notifications related to energy saving.

[0271] As shown in Figure 20, the core network node / function may also include an AI / ML module 770. If present, the AI / ML module 770 is operable to perform any of the AI / ML-related functions of the core network node / function according to one of the methods described above. The core network node / function may be configured to train or retrain AI / ML models as described above (in response to UE mobility information fed back to the core network node / function from another node in the network, such as base station 5).

[0272] (Variations and alternative examples) As those skilled in the art will understand, several modifications and substitutions can be made to the above embodiments while still benefiting from the present disclosure as embodied therein.

[0273] While the above example illustrates the concept using an AI / ML model, it should be understood that the above method is also advantageous when the model is not an AI / ML model. Any other suitable type of model or function can be used to generate inference (e.g., decision or prediction).

[0274] For example, while terminology specific to cellular communication generations (such as 2G, 3G, 4G, 5G, and 6G) ​​may be used to refer to specific communication entities for clarity, it should be understood that the technical features described for a given entity are not limited to devices of that particular communication generation. Technical features can be implemented in any functionally equivalent communication entity, regardless of the differences in terminology used to refer to them.

[0275] In the above description, the UE and base station are described as having several separate functional components or modules for the sake of ease of understanding. These modules may thus be provided for certain applications, for example, where an existing system is modified to implement the present disclosure, but in other applications, for example, in a system designed from the outset with the features of the present invention in mind, these modules may be incorporated into the operating system or the entire code, and therefore these modules may not be identifiable as separate entities.

[0276] In the exemplary embodiments described above, several software modules have been explained. As those skilled in the art will understand, software modules may be provided in compiled or uncompiled form, and may be supplied as signals over a computer network or on a recording medium. Furthermore, some or all of the functions performed by this software may be performed using one or more dedicated hardware circuits. However, the use of software modules is preferred because it facilitates updating of the base station or UE to update the functions of the base station or UE.

[0277] Each controller may include, but is not limited to, one or more hardware-implemented computer processors, microprocessors, central processing units (CPUs), arithmetic logic units (ALUs), input / output (IO) circuits, internal memory / cache (programs and / or data), processing registers, communication buses (such as control buses, data buses, and / or address buses), direct memory access (DMA) functions, hardware or software-implemented counters, pointers, and / or timers, and any other suitable form of processing circuitry. Various other modifications will be obvious to those skilled in the art and will not be described in further detail here.

[0278] The memory shown above may be formed of volatile memory or non-volatile memory, or the memory may be formed of a combination of volatile memory and non-volatile memory.

[0279] In the above exemplary embodiments, some software modules have been described. As will be understood by those skilled in the art, software modules may be provided in compiled or non-compiled form, may be supplied as a signal via a computer network, or may be supplied on a recording medium. Further, the functions performed by some or all of this software may be performed using one or more dedicated hardware circuits. However, the use of software modules is preferred because it facilitates functional updates.

[0280] The software constituting the software module can be stored using various types of non-temporary computer-readable media or tangible storage media supplied to the computer. By way of example and not limitation, non-temporary computer-readable media or tangible storage media can include random-access memory (RAM), read-only memory (ROM), flash memory, solid-state drive (SSD) or other memory technologies, compact disk-read only memory (CD-ROM), compact disk-read / write (CD-R / W), digital versatile disk (DVD), Blu-ray Disc (registered trademark), or other optical disk storage, magnetic cassette, magnetic tape, magnetic disk storage, or other magnetic storage devices. The program may also be transmitted on a temporary computer-readable medium or a communication medium. By way of example and not limitation, temporary computer-readable media or communication media can include electrical, optical, acoustic, or other forms of propagated signals.

[0281] The base station may include a "distributed" base station having a central unit "CU" and one or more individual distributed units (DUs).

[0282] In this disclosure, user equipment (or "UE," "mobile station," "mobile device," or "wireless device") is an entity connected to a network via a wireless interface.

[0283] Please note that this disclosure is not limited to dedicated communication devices, but can be applied to any device having communication functions as described in the following paragraphs.

[0284] The terms “User Equipment” or “UE” (as used by 3GPP), “Mobile Station,” “Mobile Device,” and “Radio Device” are generally intended to be synonymous with each other and include standalone mobile stations such as terminals, cell phones, smartphones, tablets, cellular IoT devices, IoT devices, and machines. The terms “Mobile Station” and “Mobile Device” will be understood to also include devices that remain stationary for extended periods.

[0285] UE may be items of equipment for production or manufacture and / or items of energy-related machinery, such as equipment or machinery (for example, boilers, engines, turbines, solar panels, wind turbines, hydroelectric generators, thermal generators, nuclear generators, batteries, nuclear systems and / or related equipment, heavy electrical machinery, pumps including vacuum pumps, compressors, fans, blowers, hydraulic equipment, pneumatic equipment, metalworking machinery, manipulators, robots and / or their application systems, tools, molds or dies, rolls, conveying equipment, elevators, material handling equipment, textile machinery, sewing machinery, printing and / or related machinery, paper conversion machinery, chemical machinery, mining machinery and / or construction machinery and / or related equipment, machinery and / or equipment for agriculture, forestry and / or fisheries, safety and / or environmental protection equipment, tractors, precision bearings, chains, gears, power transmission equipment, lubrication equipment, valves, pipe fittings and / or application systems for any of the aforementioned equipment or machinery, etc.).

[0286] UE may be an item of transport equipment, such as (for example, transport equipment such as railway cars, automobiles, motorcycles, bicycles, trains, buses, carts, rickshaws, ships and other vessels, aircraft, rockets, satellites, drones, balloons, etc.). UE may also be an item of information and communication equipment, such as (for example, information and communication equipment such as electronic computers and related equipment, communication and related equipment, electronic components, etc.).

[0287] UE may include, for example, refrigerators, refrigerator applications, trading and / or service industry equipment items, vending machines, automated service machines, office machines or equipment, and household appliances and electronic devices (such as audio equipment, video equipment, loudspeakers, radios, televisions, microwave ovens, rice cookers, coffee machines, dishwashers, washing machines, dryers, electronic fans or related equipment, vacuum cleaners, etc.).

[0288] The UE may be an electrical application system or device, for example, such as an X-ray system, particle accelerator, radioisotope equipment, sound wave equipment, electromagnetic application equipment, power application equipment, etc.

[0289] UE may include, for example, electronic lamps, lighting fixtures, measuring instruments, analyzers, testers, or measuring or detection equipment (such as smoke detectors, human alarm sensors, motion sensors, wireless tags, etc.), watches or clocks, laboratory equipment, optical devices, medical equipment and / or systems, weapons, bladed weapons, hand tools, etc.

[0290] The UE may be, for example, a wireless-equipped personal digital assistant or related device (such as a wireless card or module designed to be attached to or inserted into another electronic device, such as a personal computer or electrical measuring instrument).

[0291] The UE may be part of a device or system that uses various wired and / or wireless communication technologies to provide the following uses, services, and solutions related to the Internet of Things (IoT).

[0292] Internet of Things (IoT) devices (or "Things") may be equipped with appropriate electronics, software, sensors, network connectivity, etc., that enable them to collect and exchange data with each other and with other communication devices. IoT devices may include automated equipment that follows software instructions stored in internal memory. IoT devices may operate without requiring human supervision or interaction. IoT devices may also remain stationary and / or inactive for extended periods. IoT devices may be implemented as part of (generally) stationary equipment. IoT devices may also be incorporated into non-stationary equipment (such as a vehicle) or attached to animals or people being monitored / tracked.

[0293] It will be understood that IoT technology can be implemented on any communication device that can connect to a communication network to send / receive data, regardless of whether such communication device is controlled by human input or by software instructions stored in memory.

[0294] It will be understood that IoT devices are sometimes called Machine-Type Communication (MTC) devices or Machine-to-Machine (M2M) communication devices. It will be understood that a UE may support one or more IoT or MTC applications. Some examples of MTC applications are listed in the table below. This list is not exhaustive and is intended to inform you of some examples of machine-type communication applications. [Table 2]

[0295] Applications, services, and solutions may include MVNO (Mobile Virtual Network Operator) services, emergency radio communication systems, PBX (Private Branch eXchange) systems, PHS / digital cordless telecommunications systems, POS (Point of Sale) systems, incoming advertising systems, MBMS (Multimedia Broadcast and Multicast Service), V2X (Vehicle to Everything) systems, train radio systems, location-related services, disaster / emergency radio communication services, community services, video streaming services, femtocell application services, VoLTE (Voice over LTE) services, billing services, wireless on-demand services, roaming services, activity monitoring services, telecommunications carrier / communication network selection services, function restriction services, PoC (Proof of Concept) services, personal information management services, ad hoc network / DTN (Delay Tolerant Networking) services, and others.

[0296] Furthermore, the aforementioned UE categories are merely examples of applications of the technical concepts and exemplary embodiments described in this document. Needless to say, these technical concepts and exemplary embodiments are not limited to the aforementioned UEs and are subject to various modifications.

[0297] Various other modifications are obvious to those skilled in the art and will not be described in further detail here.

[0298] Some or all of the aspects described above can be described as follows, but the Disclosure is not limited thereto. Some or all of the elements described in any one of the appendices may apply to various types of hardware, software, and recording means for recording software, systems, and methods. (Note 1) The first access network node cell executes an artificial intelligence or machine learning (AI / ML) model, Receiving information for handover from the first access network node to the second access network node, wherein the information for handover includes information regarding the continuity of AI / ML model execution. The process is initiated based on information regarding the continuity of AI / ML model execution, A method that is performed by user equipment (UE), including [specific details]. (Note 2) Information regarding the continuity of AI / ML execution indicates whether the second access network node supports the AI / ML model. The method described in Appendix 1. (Note 3) The AI / ML model is executed for a specific use case or feature. Information regarding the continuity of AI / ML execution indicates whether the second access network node supports a specific use case or feature of the AI / ML model. The method described in Appendix 1 or 2. (Note 4) Information regarding the continuity of AI / ML execution indicates that the AI / ML model or at least one parameter used in the AI / ML model needs to be updated for continuity. The method described in any one of the appendices 1 to 3. (Note 5) Information regarding the continuity of AI / ML execution is available. Information indicating at least one AI / ML model supported by the second access network node. Information indicating at least one AI / ML model that the second access network node proposes to use, or Information indicating the status of at least one AI / ML model supported by the second access network node. including at least one of the following: The method described in any one of the appendices 1 to 4. (Note 6) If information regarding the continuity of AI / ML execution indicates that the second access network node does not support the AI / ML model, initiate the procedure to use a different AI / ML model. The method described in Appendix 2. (Note 7) The procedure is, Receiving another AI / ML model from the first access network node, or To ensure the continuity of AI / ML model execution, switch to another AI / ML model stored in the UE. including at least one of the following: The method described in Appendix 6. (Note 8) If the continuity information for AI / ML execution indicates that the second access network node does not support the AI / ML model, the AI / ML model will not be removed from the UE. The method described in Appendix 6 or 7. (Note 9) The procedure for using a different AI / ML model is initiated before the handover. The method described in any one of the appendices 6 to 8. (Note 10) To transmit preference information for at least one AI / ML model to the second access network node for use in ensuring the continuity of AI / ML model execution. The method described in any one of the appendices 1 to 9, including the method described in any one of the appendices 1 to 9. (Note 11) If the second access network node does not have at least one AI / ML model indicated by preference information, then at least one AI / ML model is transmitted from the first access network node or UE. The method described in Appendix 10. (Note 12) If the second access network node does not have at least one AI / ML model indicated by preference information, send a notification to the second access network node to cause the second access network node to receive at least one AI / ML model from a server or core network node that has at least one AI / ML model. The method described in Appendix 10, including the method described in Appendix 10. (Note 13) During handover, the first access network node receives at least one AI / ML model transmitted by the second access network node. The method described in any one of the appendices 1 to 12, including the method described in any one of the appendices 1 to 12. (Note 14) Receiving at least one AI / ML model, Radio Resource Control (RRC) message, or User plane data It is carried out via at least one of the following: The method described in Appendix 13. (Note 15) If at least one AI / ML model has not been received by the time the handover is complete, the UE will discard a portion of at least one AI / ML model that it has received. The method described in Appendix 13 or 14, including the method described in Appendix 13 or 14. (Note 16) If at least one AI / ML model has not been received by the time the handover is complete, the UE will receive at least one portion of the AI / ML model that has not been received from a second access network node. The method described in Appendix 13 or 14, including the method described in Appendix 13 or 14. (Note 17) The UE sends status information to the second access network node, indicating a portion of at least one AI / ML model received by the UE. Includes, The reception of at least one portion of an AI / ML model that the UE has not yet received is performed based on status information. The method described in Appendix 16. (Note 18) Status information indicating at least one portion of an AI / ML model sent from the first access network node to the UE is transmitted from the first access network node to the second access network node. The reception of at least one portion of an AI / ML model that the UE has not yet received is performed based on status information. The method described in Appendix 16. (Note 19) Information regarding the continuity of AI / ML model execution is transmitted from the second access network node to the first access network node. The method described in any one of the appendices 1 to 18. (Note 20) Information regarding the continuity of AI / ML model execution is available. Handover request acknowledgment message, Inter-base station interface setup response message, or Access network node configuration update acknowledgment message In at least one of these, a signal is transmitted from the second access network node to the first access network node. The method described in Appendix 19. (Note 21) Information regarding the continuity of AI / ML model execution is transmitted when the UE sends model information about the running AI / ML model from the first access network node to the second access network node. The method described in Appendix 19 or 20. (Note 22) Model information, AI / ML model identity, Information indicating the version of the AI / ML model. Information indicating the area corresponding to the AI / ML model, or Information showing the AI / ML models selected by the UE for each use case or feature. including at least one of the following: The method described in Appendix 21. (Note 23) Model information, Handover request message, Inter-base station interface setup request message, or Access network node configuration update message Included in at least one of the following: The method described in Appendix 21 or 22. (Note 24) Model information is included in the handover request message. The handover request message causes the second access network node to notify the core network node or server whether or not an AI / ML model update is required. When a core network node or server requests an update to an AI / ML model, the updated AI / ML model is sent from the core network node or server to the UE. The method described in Appendix 23. (Note 25) Model information is included in the inter-base station interface setup request message or the access network node configuration update message. Model information, Capability information indicating support for AI / ML models for each use case or feature, or Information showing supported submission methods for submitting AI / ML models. including at least one of the following: The method described in Appendix 21 or 22. (Note 26) Information regarding the continuity of AI / ML model execution is available. AI / ML model identity, Information indicating the version of the AI / ML model. Information indicating the area corresponding to the AI / ML model, or Information showing the AI / ML models selected by the UE for each use case or feature. including at least one of the following: The method described in any one of the appendices 1 to 25. (Note 27) The handover information from the first access network node to the second access network node is transmitted to user equipment (UE) running an artificial intelligence or machine learning (AI / ML) model in a cell of the first access network node, wherein the transmitted handover information includes information regarding the continuity of AI / ML model execution. Includes, Information regarding the continuity of AI / ML model execution triggers the UE to initiate the process for ensuring the continuity of AI / ML model execution. A method performed by the first access network node. (Note 28) Information regarding the continuity of AI / ML execution indicates whether the second access network node supports the AI / ML model. The method described in Appendix 27. (Note 29) The AI / ML model is executed for a specific use case or feature. Information regarding the continuity of AI / ML execution indicates whether the second access network node supports a specific use case or feature of the AI / ML model. The method described in Appendix 27 or 28. (Note 30) Information regarding the continuity of AI / ML execution indicates that the AI / ML model or at least one parameter used in the AI / ML model needs to be updated for continuity. The method described in any one of the appendices 27 to 29. (Note 31) Information regarding the continuity of AI / ML execution is available. Information indicating at least one AI / ML model supported by the second access network node. Information indicating at least one AI / ML model that the second access network node proposes to use, or Information indicating the status of at least one AI / ML model supported by the second access network node. including at least one of the following: The method described in any one of the appendices 27 to 30. (Note 32) If information regarding the continuity of AI / ML execution indicates that the second access network node does not support the AI / ML model, initiate the procedure to use a different AI / ML model. The method described in Appendix 28. (Note 33) The procedure is, Sending another AI / ML model to the user's device, or To ensure continuity in executing AI / ML models, the UE is instructed to switch to another AI / ML model stored within the UE. including at least one of the following: The method described in Appendix 32. (Note 34) If the continuity information for AI / ML execution indicates that the second access network node does not support the AI / ML model, the AI / ML model will not be removed from the UE. The method described in Appendix 32 or 33. (Note 35) The procedure for using a different AI / ML model is initiated before the handover. The method described in any one of the appendices 32 to 34. (Note 36) If the second access network node does not have at least one AI / ML model for use in the continuity of execution of the AI / ML model indicated by the UE, then transmit at least one AI / ML model to the second access network node. The method described in any one of the appendices 27 to 35, including the method described in any one of the appendices. (Note 37) If the second access network node does not have at least one AI / ML model to be used for the continuity of execution of the AI / ML model as indicated by the UE, send a notification to the second access network node to cause the second access network node to receive at least one AI / ML model from the server or core network node that has at least one AI / ML model. The method described in any one of the appendices 27 to 35, including the method described in any one of the appendices. (Note 38) During handover, send at least one AI / ML model transmitted by the second access network node to the UE. The method described in any one of the appendices 27 to 37, including the method described in any one of the appendices 27 to 37. (Note 39) At least one AI / ML model must be submitted. Radio Resource Control (RRC) message, or User plane data It is carried out via at least one of the following: The method described in Appendix 38. (Note 40) If the transmission of at least one AI / ML model is not completed by the time the handover is finished, send information to the second access network node indicating that the transmission of at least one AI / ML model will not be completed by the time the handover is finished. The method described in Appendix 38 or 39, including the method described in Appendix 38 or 39. (Note 41) Information indicating that the transmission of at least one AI / ML model was not completed by the time the handover was finished, The UE receives a portion of at least one AI / ML model, or At least one part of an AI / ML model that UE has not received Includes information indicating, The method described in Appendix 40. (Note 42) The second access network node receives information regarding the continuity of AI / ML model execution. The method described in any one of the appendices 27 to 41, including the method described in any one of the appendices. (Note 43) Information regarding the continuity of AI / ML model execution is available. Handover request acknowledgment message, Inter-base station interface setup response message, or Access network node configuration update acknowledgment message Included in at least one of the following: The method described in Appendix 42. (Note 44) Send model information about the AI / ML model being run by the UE to the second access network node. Includes, Information regarding the continuity of AI / ML model execution is received based on the transmission of model information. The method described in Appendix 42 or 43. (Note 45) Model information, AI / ML model identity, Information indicating the version of the AI / ML model. Information indicating the area corresponding to the AI / ML model, or Information showing the AI / ML models selected by the UE for each use case or feature. including at least one of the following: The method described in Appendix 44. (Note 46) Model information, Handover request message, Inter-base station interface setup request message, or Access network node configuration update message Included in at least one of the following: The method described in Appendix 44 or 45. (Note 47) Model information is included in the handover request message. The handover request message causes the second access network node to notify the core network node or server whether or not an AI / ML model update is required. When a core network node or server requests an update to an AI / ML model, the updated AI / ML model is sent from the core network node or server to the UE. The method described in Appendix 46. (Note 48) Model information is included in the inter-base station interface setup request message or the access network node configuration update message. Model information, Capability information indicating support for AI / ML models for each use case or feature, or Information showing supported submission methods for submitting AI / ML models. including at least one of the following: The method described in Appendix 44 or 45. (Note 49) Information regarding the continuity of AI / ML model execution is available. AI / ML model identity, Information indicating the version of the AI / ML model. Information indicating the area corresponding to the AI / ML model, or Information showing the AI / ML models selected by the UE for each use case or feature. including at least one of the following: The method described in any one of the appendices 27 to 48. (Note 50) The first access network node transmits information for handover from the first access network node to the second access network node, wherein the information for handover includes information regarding the continuity of execution of the artificial intelligence or machine learning (AI / ML) model by user equipment (UE) in the cell of the first access network node. Includes, Information regarding the continuity of AI / ML model execution is sent to the UE, which then initiates the process for ensuring the continuity of AI / ML model execution. A method performed by the second access network node. (Note 51) Information regarding the continuity of AI / ML execution indicates whether the second access network node supports the AI / ML model. The method described in Appendix 50. (Note 52) The AI / ML model is executed for a specific use case or feature. Information regarding the continuity of AI / ML execution indicates whether the second access network node supports a specific use case or feature of the AI / ML model. The method described in Appendix 50 or 51. (Note 53) Information regarding the continuity of AI / ML execution indicates that the AI / ML model or at least one parameter used in the AI / ML model needs to be updated for continuity. The method described in any one of the appendices 50 to 52. (Note 54) Information regarding the continuity of AI / ML execution is available. Information indicating at least one AI / ML model supported by the second access network node. Information indicating at least one AI / ML model that the second access network node proposes to use, or Information indicating the status of at least one AI / ML model supported by the second access network node. including at least one of the following: The method described in any one of the appendices 50 to 53. (Note 55) If AI / ML execution continuity information indicates that the second access network node does not support the AI / ML model, the AI / ML execution continuity information prompts the UE to initiate a procedure to use a different AI / ML model. The method described in Appendix 51. (Note 56) The procedure is, Sending another AI / ML model from the first access network node to the UE, or UE allows switching to another AI / ML model stored in the UE for continuity of AI / ML model execution. including at least one of the following: The method described in Appendix 55. (Note 57) If the continuity information for AI / ML execution indicates that the second access network node does not support the AI / ML model, the AI / ML model will not be removed from the UE. The method described in Appendix 55 or 56. (Note 58) The procedure for using a different AI / ML model is initiated before the handover. The method described in any one of the appendices 55 to 57. (Note 59) To receive preference information for at least one AI / ML model from the UE for use in ensuring the continuity of AI / ML model execution. The method described in any one of the appendices 50 to 58, including the method described in any one of the appendices 50 to 58. (Note 60) If the second access network node does not have at least one AI / ML model indicated by preference information, request the first access network node or UE to send at least one AI / ML model to the second access network node. The method described in Appendix 59, including the method described therein. (Note 61) If the second access network node does not have at least one AI / ML model indicated by preference information, Receiving notifications from UE, Based on the notification, receive at least one AI / ML model from a server or core network node owned by at least one AI / ML model, The method described in Appendix 59, including the method described therein. (Note 62) During handover, send at least one AI / ML model to the UE via the first access network node. The method described in any one of the appendices 50 to 61, including the method described in any one of the appendices 50 to 61. (Note 63) At least one AI / ML model must be submitted. Radio Resource Control (RRC) message, or User plane data It is carried out via at least one of the following: The method described in Appendix 62. (Note 64) If at least one AI / ML model has not been sent by the time the handover is complete, At least one AI / ML model, or At least one part of an AI / ML model that UE has not received Send it directly to the UE The method described in Appendix 62 or 63, including the method described in Appendix 62 or 63. (Note 65) The UE receives status information indicating a portion of at least one AI / ML model that the UE has received. Includes, The transmission of at least one AI / ML model portion that the UE has not received will be performed based on status information. The method described in Appendix 64. (Note 66) The first access network node receives status information indicating a portion of at least one AI / ML model sent to the UE. Includes, The transmission of at least one AI / ML model portion that the UE has not received will be performed based on status information. The method described in Appendix 64. (Note 67) Information regarding the continuity of AI / ML model execution is available. Handover request acknowledgment message, Inter-base station interface setup response message, or Access network node configuration update acknowledgment message Included in at least one of the following: The method described in any one of the appendices 50 to 66. (Note 68) The first access network node receives model information about the AI / ML model being executed by the UE. Includes, Information regarding the continuity of AI / ML model execution is transmitted based on model information. The method described in any one of the appendices 50 to 67. (Note 69) Model information, AI / ML model identity, Information indicating the version of the AI / ML model. Information indicating the area corresponding to the AI / ML model, or Information showing the AI / ML models selected by the UE for each use case or feature. including at least one of the following: The method described in Appendix 68. (Note 70) Model information, Handover request message, Inter-base station interface setup request message, or Access network node configuration update message Included in at least one of the following: The method described in Appendix 68 or 69. (Note 71) Model information is included in the handover request message. The method is Notifying core network nodes or servers whether or not an AI / ML model needs to be updated, When a core network node or server requests an update to an AI / ML model, Receiving updated AI / ML models from core network nodes or servers, Sending the updated AI / ML model to the UE, including, The method described in Appendix 70. (Note 72) Model information is included in the inter-base station interface setup request message or the access network node configuration update message. Model information, Capability information indicating support for AI / ML models for each use case or feature, or Information showing supported submission methods for submitting AI / ML models. including at least one of the following: The method described in Appendix 68 or 69. (Note 73) Information regarding the continuity of AI / ML model execution is available. AI / ML model identity, Information indicating the version of the AI / ML model. Information indicating the area corresponding to the AI / ML model, or Information showing the AI / ML models selected by the UE for each use case or feature. including at least one of the following: The method described in any one of the appendices 50 to 72. (Note 74) A means for executing an artificial intelligence or machine learning (AI / ML) model in a cell of the first access network node, A means for receiving information for handover from a first access network node to a second access network node, wherein the handover information includes information regarding the continuity of execution of the AI / ML model. A means of initiating a process based on information regarding the continuity of AI / ML model execution, User equipment (UE) that includes these features. (Note 75) Means for transmitting information for handover from a first access network node to a second access network node to user equipment (UE) running an artificial intelligence or machine learning (AI / ML) model in a cell of a first access network node, wherein the information for handover includes information regarding the continuity of execution of the AI / ML model. Equipped with, Information regarding the continuity of AI / ML model execution triggers the UE to initiate the process for ensuring the continuity of AI / ML model execution. The first access network node. (Note 76) Means for transmitting information for a handover from a first access network node to a second access network node, wherein the information for the handover includes information regarding the continuity of execution of an artificial intelligence or machine learning (AI / ML) model by user equipment (UE) in a cell of the first access network node. Equipped with, Information regarding the continuity of AI / ML model execution is sent to the UE, which then initiates the process for ensuring the continuity of AI / ML model execution. The second access network node.

[0299] This application is based on and claims priority to UK Patent Application No. 2302758.4, filed on 24 February 2023, the disclosure of which is incorporated herein by reference in its entirety. [Explanation of Symbols]

[0300] 1. Communication System 3. User equipment 5, 5-1, 5-2 RAN Nodes (Base Stations, RAN Equipment) 7 Core Network 9 cells 10 Control Plane Functions 11. User Plane Functions 20 External data network 41 Data Collection Function 43 Model Training Functions 45 Model Inference Functions 47 Actors 50 Distributed Units 60 Central Unit 151 AI / ML Servers 180 First cell 181 Second cell 182 Third Cell 310 Transceiver Circuit 330 Antenna 350 User Interfaces 370 Controller 390 memory 410 Operating Systems 430 Communication control module 450 AI / ML Modules 451 Transceiver Circuit 453 RU interface 454 CU interface 457 Controller 459 memory 461 Operating Systems 463 Communication control module 465 F1 module 468 DU-RU Module 472 DU Management Module 473 UE Profile Management Module 475 Mobility Module 510 Transceiver Circuit 530 Antenna 550 Core Network Interfaces 551 Transceiver Circuit 554 DU Interface 555 CU Interface 557 Controller 559 memory 561 Operating Systems 563 Communication control module 565 F1 module 566 E1 module 568 N2 module 569 N3 module 570 Controller 571 CU-UP Management Module 572 CU-CP Management Module 573 UE Profile Management Module 575 Mobility Module 590 memory 610 Operating Systems 630 Communication control module 650 AI / ML Modules 710 Transceiver Circuit 720 Network Interfaces 730 Controller 740 memory 750 Operating Systems 760 Communication Control Module 770 AI / ML Module< / cu> < / du>

Claims

1. Executing an artificial intelligence or machine learning (AI / ML) model in the cell of the first access network node, Receiving information from the first access network node for handover to the second access network node, wherein the information for handover includes information regarding the continuity of the execution of the AI / ML model. The process is initiated based on the information regarding the continuity of the execution of the AI / ML model, A method that is performed by user equipment (UE), including [specific details omitted].

2. The information relating to the continuity of the execution of the AI / ML indicates whether the second access network node supports the AI / ML model. The method according to claim 1.

3. The execution of the AI / ML model is performed for a specific use case or feature. The information relating to the continuity of the execution of the AI / ML indicates whether the second access network node supports the specific use case or feature of the AI / ML model. The method according to claim 1 or 2.

4. The information relating to the continuity of the execution of the AI / ML indicates that the AI / ML model or at least one parameter used in the AI / ML model needs to be updated for the sake of continuity. The method according to any one of claims 1 to 3.

5. The information relating to the continuity of the execution of the AI / ML is Information indicating at least one AI / ML model supported by the second access network node, Information indicating at least one AI / ML model that the second access network node proposes to use, or Information indicating the status of each of the at least one AI / ML models supported by the second access network node. including at least one of the following: The method according to any one of claims 1 to 4.

6. If the information relating to the continuity of the execution of the AI / ML indicates that the second access network node does not support the AI / ML model, then a procedure for using a different AI / ML model is initiated. The method according to claim 2.

7. The above procedure, Receiving the other AI / ML model from the first access network node, or To ensure the continuity of the execution of the AI / ML model, switch to another AI / ML model stored in the UE. including at least one of the following: The method according to claim 6.

8. If the information regarding the continuity of the execution of the AI / ML indicates that the second access network node does not support the AI / ML model, the AI / ML model is not removed from the UE. The method according to claim 6 or 7.

9. The initiation of the procedure for using the other AI / ML model is performed before the handover. The method according to any one of claims 6 to 8.

10. Transmit to the second access network node the preference information of at least one AI / ML model for use in ensuring the continuity of the execution of the AI / ML model. The method according to any one of claims 1 to 9, including the method described in any one of claims 1 to 9.

11. If the second access network node does not have the at least one AI / ML model indicated by the preference information, the at least one AI / ML model is transmitted from the first access network node or the UE. The method according to claim 10.

12. If the second access network node does not have the at least one AI / ML model indicated by the preference information, send a notification to the second access network node to cause the second access network node to receive the at least one AI / ML model from the server or core network node that has the at least one AI / ML model. The method according to claim 10, including the method described in claim 10.

13. During the handover, the first access network node receives at least one AI / ML model transmitted by the second access network node. The method according to any one of claims 1 to 12, including the method described in any one of claims 1 to 12.

14. The reception of the at least one AI / ML model is Radio Resource Control (RRC) messages, or User plane data This is carried out via at least one of the following: The method according to claim 13.

15. If the reception of the at least one AI / ML model is not completed by the time the handover is completed, the UE discards a portion of the at least one AI / ML model that it has received. The method according to claim 13 or 14, including the method described in claim 13 or 14.

16. If the reception of at least one AI / ML model is not completed by the time the handover is completed, the UE shall receive the portion of the at least one AI / ML model that has not been received from the second access network node. The method according to claim 13 or 14, including the method described in claim 13 or 14.

17. To transmit status information to the second access network node, which indicates a portion of the at least one AI / ML model received by the UE. Includes, The reception of the portion of the at least one AI / ML model that the UE has not received is performed based on the status information. The method according to claim 16.

18. Status information indicating a portion of the at least one AI / ML model transmitted by the first access network node to the UE is transmitted from the first access network node to the second access network node. The reception of the portion of the at least one AI / ML model that the UE has not received is performed based on the status information. The method according to claim 16.

19. The information relating to the continuity of the execution of the AI / ML model is transmitted from the second access network node to the first access network node. The method according to any one of claims 1 to 18.

20. The information relating to the continuity of the execution of the AI / ML model is Handover request acknowledgment message, Inter-base station interface setup response message, or Access network node configuration update acknowledgment message In at least one of the following, a transmission from the second access network node to the first access network node is transmitted: The method according to claim 19.

21. The information relating to the continuity of execution of the AI / ML model is transmitted when the UE transmits model information relating to the AI / ML model that is being executed from the first access network node to the second access network node. The method according to claim 19 or 20.

22. The aforementioned model information, The identity of the aforementioned AI / ML model, Information indicating the version of the aforementioned AI / ML model, Information indicating the area corresponding to the aforementioned AI / ML model, or Information indicating each AI / ML model selected by the UE for each use case or feature. including at least one of the following: The method according to claim 21.

23. The aforementioned model information, Handover request message, Inter-base station interface setup request message, or Access network node configuration update message Included in at least one of the following: The method according to claim 21 or 22.

24. The aforementioned model information is included in the handover request message. The handover request message causes the second access network node to notify the core network node or server whether or not the AI / ML model needs to be updated. When the core network node or the server requests an update to the AI / ML model, the updated AI / ML model is transmitted from the core network node or the server to the UE. The method according to claim 23.

25. The aforementioned model information is included in the base station interface setup request message or the access network node configuration update message, The aforementioned model information, Capability information indicating support for the aforementioned AI / ML model for each use case or feature, or Information indicating supported transmission methods for transmitting the aforementioned AI / ML model. including at least one of the following: The method according to claim 21 or 22.

26. The information relating to the continuity of the execution of the AI / ML model is The identity of the aforementioned AI / ML model, Information indicating the version of the aforementioned AI / ML model, Information indicating the area corresponding to the aforementioned AI / ML model, or Information indicating each AI / ML model selected by the UE for each use case or feature. including at least one of the following: The method according to any one of claims 1 to 25.

27. Transmitting information for a handover from the first access network node to a second access network node to user equipment (UE) running an artificial intelligence or machine learning (AI / ML) model in a cell of the first access network node, wherein the information for the handover includes information regarding the continuity of the execution of the AI / ML model. Includes, The information relating to the continuity of the execution of the AI / ML model causes the UE to initiate a process for the continuity of the execution of the AI / ML model. A method performed by the first access network node.

28. The information relating to the continuity of the execution of the AI / ML indicates whether the second access network node supports the AI / ML model. The method according to claim 27.

29. The execution of the AI / ML model is performed for a specific use case or feature. The information relating to the continuity of the execution of the AI / ML indicates whether the second access network node supports the specific use case or feature of the AI / ML model. The method according to claim 27 or 28.

30. The information relating to the continuity of the execution of the AI / ML indicates that the AI / ML model or at least one parameter used in the AI / ML model needs to be updated for the sake of continuity. The method according to any one of claims 27 to 29.

31. The information relating to the continuity of the execution of the AI / ML is Information indicating at least one AI / ML model supported by the second access network node, Information indicating at least one AI / ML model that the second access network node proposes to use, or Information indicating the status of each of the at least one AI / ML models supported by the second access network node. including at least one of the following: The method according to any one of claims 27 to 30.

32. If the information relating to the continuity of the execution of the AI / ML indicates that the second access network node does not support the AI / ML model, then a procedure for using a different AI / ML model is initiated. The method according to claim 28.

33. The above procedure, Transmitting the other AI / ML model to the user device, or To cause the UE to switch to another AI / ML model stored in the UE in order to ensure the continuity of the execution of the AI / ML model, including at least one of the following: The method according to claim 32.

34. If the information regarding the continuity of the execution of the AI / ML indicates that the second access network node does not support the AI / ML model, the AI / ML model is not removed from the UE. The method according to claim 32 or 33.

35. The initiation of the procedure for using the other AI / ML model is performed before the handover. The method according to any one of claims 32 to 34.

36. If the second access network node does not have at least one AI / ML model for use in the continuity of the execution of the AI / ML model indicated by the UE, transmit the at least one AI / ML model to the second access network node. The method according to any one of claims 27 to 35, including the method described in any one of claims 27 to 35.

37. If the second access network node does not have at least one AI / ML model used for the continuity of the execution of the AI / ML model as indicated by the UE, then send a notification to the second access network node to cause the second access network node to receive the at least one AI / ML model from the server or core network node that has the at least one AI / ML model. The method according to any one of claims 27 to 35, including the method described in any one of claims 27 to 35.

38. During the handover, transmit at least one AI / ML model transmitted by the second access network node to the UE. The method according to any one of claims 27 to 37, including the method described in any one of claims 27 to 37.

39. The transmission of the at least one AI / ML model is Radio Resource Control (RRC) messages, or User plane data This is carried out via at least one of the following: The method according to claim 38.

40. If the transmission of the at least one AI / ML model is not completed by the time the handover is finished, information indicating that the transmission of the at least one AI / ML model is not completed by the time the handover is finished is transmitted to the second access network node. The method according to claim 38 or 39, including the method described in claim 38 or 39.

41. Information indicating that the transmission of at least one AI / ML model has not been completed by the time of the handover completion, The UE receives a portion of the at least one AI / ML model, or A portion of the at least one AI / ML model that the UE has not received Includes information indicating, The method according to claim 40.

42. Receiving the information regarding the continuity of the execution of the AI / ML model from the second access network node. The method according to any one of claims 27 to 41, including the method described in any one of claims 27 to 41.

43. The information relating to the continuity of the execution of the AI / ML model is Handover request acknowledgment message, Inter-base station interface setup response message, or Access network node configuration update acknowledgment message Included in at least one of the following: The method according to claim 42.

44. To transmit model information relating to the AI / ML model being executed by the UE to the second access network node. Includes, The reception of the information relating to the continuity of the execution of the AI / ML model is performed based on the transmission of the model information. The method according to claim 42 or 43.

45. The aforementioned model information, The identity of the aforementioned AI / ML model, Information indicating the version of the aforementioned AI / ML model, Information indicating the area corresponding to the aforementioned AI / ML model, or Information indicating the AI / ML model selected by the UE for each use case or feature. including at least one of the following: The method according to claim 44.

46. The aforementioned model information, Handover request message, Inter-base station interface setup request message, or Access network node configuration update message Included in at least one of the following: The method according to claim 44 or 45.

47. The aforementioned model information is included in the handover request message. The handover request message causes the second access network node to notify the core network node or server whether or not the AI / ML model needs to be updated. When the core network node or the server requests an update to the AI / ML model, the updated AI / ML model is transmitted from the core network node or the server to the UE. The method according to claim 46.

48. The aforementioned model information is included in the base station interface setup request message or the access network node configuration update message, The aforementioned model information, Capability information indicating support for the aforementioned AI / ML model for each use case or feature, or Information indicating supported transmission methods for transmitting the aforementioned AI / ML model. including at least one of the following: The method according to claim 44 or 45.

49. The information relating to the continuity of the execution of the AI / ML model is The identity of the aforementioned AI / ML model, Information indicating the version of the aforementioned AI / ML model, Information indicating the area corresponding to the aforementioned AI / ML model, or Information indicating each AI / ML model selected by the UE for each use case or feature. including at least one of the following: The method according to any one of claims 27 to 48.

50. Transmitting information to a first access network node for a handover from the first access network node to a second access network node, wherein the information for the handover includes information regarding the continuity of execution by user equipment (UE) of an artificial intelligence or machine learning (AI / ML) model in a cell of the first access network node. Includes, The information relating to the continuity of the execution of the AI / ML model is transmitted to the UE, causing the UE to initiate a process for the continuity of the execution of the AI / ML model. A method executed by a second access network node.

51. The information relating to the continuity of the execution of the AI / ML indicates whether the second access network node supports the AI / ML model. The method according to claim 50.

52. The execution of the AI / ML model is performed for a specific use case or feature. The information relating to the continuity of the execution of the AI / ML indicates whether the second access network node supports the specific use case or feature of the AI / ML model. The method according to claim 50 or 51.

53. The information relating to the continuity of the execution of the AI / ML indicates that the AI / ML model or at least one parameter used in the AI / ML model needs to be updated for the sake of continuity. The method according to any one of claims 50 to 52.

54. The information relating to the continuity of the execution of the AI / ML is Information indicating at least one AI / ML model supported by the second access network node, Information indicating at least one AI / ML model that the second access network node proposes to use, or Information indicating the status of each of the at least one AI / ML models supported by the second access network node. including at least one of the following: The method according to any one of claims 50 to 53.

55. If the information relating to the continuity of the execution of the AI / ML indicates that the second access network node does not support the AI / ML model, the information relating to the continuity of the execution of the AI / ML causes the UE to initiate a procedure to use an AI / ML model other than the AI / ML model. The method according to claim 51.

56. The above procedure, Transmitting the other AI / ML model from the first access network node to the UE, or The UE switches to another AI / ML model stored in the UE in order to ensure the continuity of the execution of the AI / ML model. including at least one of the following: The method according to claim 55.

57. If the information regarding the continuity of the execution of the AI / ML indicates that the second access network node does not support the AI / ML model, the AI / ML model is not removed from the UE. The method according to claim 55 or 56.

58. The initiation of the procedure for using the other AI / ML model is performed before the handover. The method according to any one of claims 55 to 57.

59. To receive preference information for at least one AI / ML model from the UE for use in the continuity of the execution of the AI / ML model. The method according to any one of claims 50 to 58, including the method described in any one of claims 50 to 58.

60. If the second access network node does not have the at least one AI / ML model indicated by the preference information, request the first access network node or the UE to transmit the at least one AI / ML model to the second access network node. The method according to claim 59, including the method described in claim 59.

61. If the second access network node does not have the at least one AI / ML model indicated by the preference information, Receiving notifications from the aforementioned UE, Based on the aforementioned notification, the at least one AI / ML model is received from a server or core network node owned by the at least one AI / ML model, The method according to claim 59, including the method described in claim 59.

62. During the handover, transmit at least one AI / ML model to the UE via the first access network node. The method according to any one of claims 50 to 61, including the method described in any one of claims 50 to 61.

63. The transmission of the at least one AI / ML model is Radio Resource Control (RRC) messages, or User plane data This is carried out via at least one of the following: The method according to claim 62.

64. If the transmission of the at least one AI / ML model is not completed by the time the handover is completed, All of the aforementioned at least one AI / ML model, or A portion of the at least one AI / ML model that the UE has not received To send directly to the aforementioned UE The method according to claim 62 or 63, including the method described in claim 62 or 63.

65. The UE receives status information from the UE indicating a portion of the at least one AI / ML model that the UE has received. Includes, The transmission of the portion of the at least one AI / ML model that has not been received by the UE is performed based on the status information. The method according to claim 64.

66. This includes receiving status information indicating a portion of the at least one AI / ML model transmitted by the first access network node to the UE, The transmission of the portion of the at least one AI / ML model that has not been received by the UE is performed based on the status information. The method according to claim 64.

67. The information relating to the continuity of the execution of the AI / ML model is Handover request acknowledgment message, Inter-base station interface setup response message, or Access network node configuration update acknowledgment message Included in at least one of the following: The method according to any one of claims 50 to 66.

68. The UE receives model information relating to the AI / ML model being executed from the first access network node. Includes, The transmission of the information relating to the continuity of execution of the AI / ML model is performed based on the model information. The method according to any one of claims 50 to 67.

69. The aforementioned model information, The identity of the aforementioned AI / ML model, Information indicating the version of the aforementioned AI / ML model, Information indicating the area corresponding to the aforementioned AI / ML model, or Information indicating each AI / ML model selected by the UE for each use case or feature. including at least one of the following: The method according to claim 68.

70. The aforementioned model information, Handover request message, Inter-base station interface setup request message, or Access network node configuration update message Included in at least one of the following: The method according to claim 68 or 69.

71. The aforementioned model information is included in the handover request message. The method described above is Notifying the core network node or server whether or not the aforementioned AI / ML model needs to be updated, If the core network node or the server requests an update to the AI / ML model, Receiving updated AI / ML models from the core network node or the server, The updated AI / ML model is transmitted to the UE, including, The method according to claim 70.

72. The aforementioned model information is included in the base station interface setup request message or the access network node configuration update message, The aforementioned model information, Capability information indicating support for the aforementioned AI / ML model for each use case or feature, or Information indicating supported transmission methods for transmitting the aforementioned AI / ML model. including at least one of the following: The method according to claim 68 or 69.

73. The information relating to the continuity of the execution of the AI / ML model is The identity of the aforementioned AI / ML model, Information indicating the version of the aforementioned AI / ML model, Information indicating the area corresponding to the aforementioned AI / ML model, or Information indicating each AI / ML model selected by the UE for each use case or feature. including at least one of the following: The method according to any one of claims 50 to 72.

74. A means for executing an artificial intelligence or machine learning (AI / ML) model in a cell of the first access network node, A means for receiving information for a handover from the first access network node to a second access network node, wherein the information for the handover includes information regarding the continuity of the execution of the AI / ML model. Means for initiating a process based on the information relating to the continuity of the execution of the AI / ML model, User equipment (UE) that includes these features.

75. Means for transmitting information for a handover from the first access network node to a second access network node to user equipment (UE) running an artificial intelligence or machine learning (AI / ML) model in a cell of a first access network node, wherein the information for the handover includes information regarding the continuity of the execution of the AI / ML model. Equipped with, The information relating to the continuity of the execution of the AI / ML model causes the UE to initiate a process for the continuity of the execution of the AI / ML model. The first access network node.

76. Means for transmitting information to a first access network node for a handover from the first access network node to a second access network node, wherein the information for the handover includes information regarding the continuity of execution by user equipment (UE) of an artificial intelligence or machine learning (AI / ML) model in a cell of the first access network node. Equipped with, The information relating to the continuity of the execution of the AI / ML model is transmitted to the UE, causing the UE to initiate a process for the continuity of the execution of the AI / ML model. The second access network node.

Images