Mobile device, access network node, and method

Abstract

The present disclosure relates to a method implemented by an access network node, the method including: transmitting to a user equipment, UE, notification of features for implementation in a cell of the access network node, the features being implemented using a corresponding model for generating determination, prediction, or output parameters; receiving a request for the model from the UE; and transmitting to the UE model transmission information including notification of one or more communication resources to be used by the UE to receive the model, or at least one of the models.

Description

[Technical Field]

[0001] This disclosure relates to communication systems. This disclosure has particular, but not exclusive, relevance to wireless communication systems and devices therein that operate in accordance with 3rd Generation Partnership Project (3GPP®) standards or equivalents or derivatives thereof (including LTE-Advanced, next-generation or 5G networks, future generations, and beyond). This disclosure has particular, but not necessarily exclusive, relevance to artificial intelligence and machine learning (AI / ML) models used in "new radio" systems (also known as "next-generation" systems) and similar systems. [Background technology]

[0002] Recent developments in 3GPP standards are referred to as the Long-Term Evolution (LTE) of the Evolved Packet Core (EPC) network and the Evolved UMTS Terrestrial Radio Access Network (E-UTRAN), commonly referred to as "4G." Additionally, the terms "5G" and "new radio" (NR) refer to evolving communications technologies expected to support a variety of applications and services. Various details of 5G networks are described, for example, in the "NGMN 5G White Paper" V1.0 by the Next Generation Mobile Network (NGMN) Alliance, available 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 Network.

[0003] Under 3GPP standards, a NodeB (or eNB in ​​LTE, gNB in ​​5G) is a Radio Access Network (RAN) node (or simply "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 (R)AN node, base station, or access network node to refer to any such access node.

[0004] Some additional developments in 3GPP relate to 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 a UE's path based on the UE's previous mobility, can be used for beam management, or can be used in ways to encode and transmit information. An AI / ML model may be hosted in a base station, and the base station may perform control of communication resources or control of the UE's state (e.g., control of UE mobility or control of the UE's radio resource control, RRC, state) based on inferences (e.g., decisions or predictions) generated using the AI / ML model. The base station may also transmit inferences generated using the model to another node in the network for use by the other node. Alternatively, an AI / ML model may be hosted in two nodes of the network, e.g., a base station and a UE. In this case, both the base station and the UE can use the model to make decisions or predictions. For example, a UE may use a model as part of its encoding process to encode (and / or compress) channel state information (CSI) for transmission to a base station, and the base station may use the same model as part of a corresponding decoding (and / or decompression) process to decode CSI received from the UE. [Prior art documents] [Non-patent literature]

[0005] [Non-Patent Document 1] NGMN Alliance, “5G White Paper”(https: / / www.ngmn.org / 5g-white-paper.html) Summary of the Invention [Problem to be solved by the invention]

[0006] There is a need for improved methods for propagating AI / ML models and related information between nodes in a communications network. For example, a relatively large amount of data may be required to transfer an AI / ML model, and a more efficient and reliable method for transferring the AI / ML model is needed. Furthermore, AI / ML models for deployment on a UE may not be stored in the base station that communicates with the UE, but may be stored on an external server. There is a need for an efficient and reliable mechanism for transferring AI / ML models from the server to nodes that use the models.

[0007] Another problem is that the transfer of the AI / ML model to the network node may be interrupted, for example, due to a radio link failure between the UE and the base station. A method is needed to mitigate such interruptions. For example, an improved method is needed when the transfer of the AI / ML model from the base station to the UE is interrupted due to a radio link failure, but the AI / ML model is no longer available at the base station after the radio link is restored.

[0008] In some implementations, a base station may need to determine which model is being used by a UE within the base station's cell. For example, in the case of so-called "two-sided" models where the UE and base station generate joint inferences (e.g., for encoding and decoding) using a model provided in the UE and a corresponding model provided in the base station, synchronization between the model version in the UE and the model version in the base station may be required. Improved methods for monitoring and controlling AI / ML models used in UEs are needed.

[0009] Although the UE may determine to retrieve and execute an AI / ML model or may be instructed by the network to retrieve and execute an AI / ML model, there is also the problem that in some scenarios the UE may not have sufficient memory or processing resources to store or execute the model. More generally, there is a need for improved methods for enabling more efficient and reliable transmission of AI / ML models between nodes in a communications network, as well as control of the use of AI / ML models by entities within the network.

[0010] The present disclosure aims to provide apparatus and methods that at least partially address the above needs and / or problems. [Means for solving the problem]

[0011] In a first aspect, the present invention provides a method implemented by an access network node, the method comprising: transmitting to a user equipment (UE) an indication of features for implementation in a cell of the access network node, the features being implemented using corresponding models for generating decision, prediction or output parameters; receiving a request for the models from the UE; and transmitting to the UE model transmission information comprising an indication of one or more communication resources to be used by the UE to receive the models, or at least one of the models.

[0012] The model may be an artificial intelligence or machine learning (AI / ML) model. The notification of the feature may be sent by the access network node in the system information.

[0013] The method may further include transmitting, to the UE, information indicating one or more communication resources used by the UE to request model information corresponding to the model, receiving a request for the model information from the UE, and transmitting the model information to the UE. The model information may include at least one of an indication of the identity of the model or a version number of the model.

[0014] The method may further include receiving information from the UE indicating characteristics of the UE, and determining at least one of a model to send to the UE or a configuration of the model to send to the UE based on the characteristics of the UE. The characteristics of the UE may include at least one of a capability of the UE, a type of the UE, an indication of a model supported by the UE, or an indication of a configuration of a model supported by the UE.

[0015] The method may further include transmitting at least one of an indication of the model to be transmitted to the UE, an indication of a configuration of the model to be transmitted to the UE, or a size of the model to the UE. A configuration of a model may include one or more parameters for use with the model to generate a decision, prediction, or output parameter.

[0016] The method may further include transmitting the model to the UE using the indicated communication resources.

[0017] The signaled communication resources may be for use by the UE to receive models from nodes other than the access network node. The node other than the access network node may be a server that stores the model, or a core network node.

[0018] The advertised communication resources may include at least one of a network address of a node other than the access network node, or a configuration of a radio bearer for receiving a model from a node other than the access network node. The request for the model may include an indication of the identity of the model requested by the UE.

[0019] The one or more communication resources used by the UE to receive the model may include at least one of a time resource or a frequency resource used by the UE to receive the model. The one or more communication resources used by the UE to receive the model may include a notification that the model should be sent to the UE after a predetermined period of time has elapsed.

[0020] When the access network node transmits the model to the UE, the access network node may transmit the model to multiple UEs, including the UE, in a broadcast or multicast transmission. If the access network node transmits the model to the UE, the method may further include receiving an indication from the UE of whether the model has been received at the UE.

[0021] The method may further include determining whether the model was received at the UE based on the notification of whether the model was received at the UE, and if the access network node determines that the model was not received at the UE, retransmitting the model to the UE.

[0022] The method may include receiving a request for a model in a radio resource control (RRC) transmission and transmitting the model to a UE using the RRC transmission. The model may be an AI / ML model, and the method may include at least one of receiving a request for the model using a dedicated protocol layer for transmission of information related to the AI / ML model or transmitting the model to the UE using the dedicated protocol layer.

[0023] The model may be an AI / ML model, and the method may include receiving, in an RRC transmission, a request for the model and transmitting the model to the UE using a dedicated protocol layer for transmission of information related to the AI / ML model. The model may be an AI / ML model, and the method may include receiving a request for the model in an RRC transmission, wherein the notification indicating one or more communication resources to be used by the UE to receive the model includes notification that the UE will receive the model using an application layer protocol.

[0024] The indication of one or more communication resources to be used by the UE to receive the model may include an application layer transport address. The model may be an AI / ML model, and the notification of the one or more communication resources to be used by the UE to receive the model may include notification that the UE will receive the model using a dedicated protocol stack for transmission of information related to the AI / ML model. In another aspect, the present disclosure provides a method, performed by an access network node, comprising: receiving, from user equipment (UE), UE capability information indicating at least one of features supported by the UE, where the feature is implemented using a corresponding model for generating the determination, prediction, or output parameter, a model for generating the determination, prediction, or output parameter supported by the UE, or an indication of one of the models stored in the UE; determining, based on the UE capability information, a model to be transmitted to or activated in the UE; and transmitting, to the UE, at least one of model transmission information including an indication of one or more communication resources to be used by the UE to receive the model, the model, or a UE request to activate the model.

[0025] The model may be an artificial intelligence or machine learning (AI / ML) model. The method may further include determining, based on the UE capability information, to request further UE capability information from the UE, sending a request for the further UE capability information to the UE, receiving the further UE capability information from the UE, and determining, based on the further UE capability information, a model to be sent to the UE.

[0026] The further UE capability information may include at least one of an indication of a model version supported by the UE or an indication of one or more models stored in the UE.

[0027] The advertised communication resource or resources may be for use by the UE to receive a model from a node other than an access network node. The node other than the access network node may be a server that stores the model, or a core network node.

[0028] The advertised communication resources may include at least one of a network address of a node other than the access network node, or a configuration of a radio bearer for receiving a model from a node other than the access network node. The one or more communication resources used by the UE to receive the model may include at least one of a time resource or a frequency resource used by the UE to receive the model. If the access network node transmits the model to the UE, the method may further include receiving an indication from the UE of whether the model has been received at the UE.

[0029] The method may further include determining whether the model was received at the UE based on the notification of whether the model was received at the UE, and if the access network node determines that the model was not received at the UE, retransmitting the model to the UE.

[0030] The method may include, after transmitting the model to the UE, the UE transmitting a request for the UE to activate the model.

[0031] The method may include determining that a model is stored in the UE and transmitting a request for the UE to activate the model stored in the UE. In another aspect, the present disclosure provides a method performed by a user equipment (UE), the method including receiving, from an access network node, indication of features for implementation in a cell of the access network node, the features being implemented using a corresponding model for generating determination, prediction, or output parameters; determining to obtain the model; transmitting a request for the model to the access network node; and receiving from the access network node model transmission information including indication of one or more communication resources to be used by the UE to receive the model or at least one of the models.

[0032] The model may be an artificial intelligence or machine learning (AI / ML) model. Determining to obtain the model may include determining to obtain the model if the model is not stored in the UE.

[0033] The notification of the feature may be received from the access network node in the system information.

[0034] The method may further include receiving, from the access network node, information indicating one or more communication resources used by the UE to request model information corresponding to the model, sending a request for the model information to the access network node, and receiving the model information from the access network node. The model information may include at least one of an indication of the identity of the model and a version number of the model.

[0035] The method may further include receiving the model from the access network node using the indicated communication resources.

[0036] The notified communications resources may be for use by the UE to receive the model from a node other than the access network node, and the method includes receiving the model from the node other than the access network node. The node other than the access network node may be a server that stores the model, or a core network node.

[0037] The advertised communication resources may include at least one of a network address of a node other than the access network node, or a configuration of a radio bearer for receiving a model from a node other than the access network node.

[0038] The method may include sending a request for the model to a node other than the access network node based on the notified communications resources. The method may further include sending to the access network node a notification of whether the model was received at the UE.

[0039] In another aspect, the present disclosure provides a method performed by a user equipment (UE), the method including: transmitting to an access network node UE capability information indicating at least one of features supported by the UE, where the feature is implemented using a corresponding model for generating the determination, prediction, or output parameter, a model for generating the determination, prediction, or output parameter supported by the UE, or an indication of one of the models stored in the UE; and receiving from the access network node at least one of model transmission information including an indication of one or more communications resources to be used by the UE to receive the model, the model, or a UE request to activate the model.

[0040] If the UE receives a request for the UE to activate the model, the UE may activate the model.

[0041] The model may be an artificial intelligence or machine learning (AI / ML) model. The method may further include receiving a request for further UE capability information from the access network node, and transmitting the further UE capability information to the access network node.

[0042] The further UE capability information may include at least one of an indication of a model version supported by the UE or an indication of one or more models stored in the UE.

[0043] The advertised communication resource or resources may be for use by the UE to receive a model from a node other than an access network node. The node other than the access network node may be a server that stores the model, or a core network node.

[0044] The advertised communication resources may include at least one of a network address of a node other than the access network node, or a configuration of a radio bearer for receiving a model from a node other than the access network node.

[0045] The one or more communication resources used by the UE to receive the model may include at least one of a time resource or a frequency resource used by the UE to receive the model. If the UE receives the model from an access network node, the method may further include sending an indication to the access network node of whether the model has been received at the UE.

[0046] In another aspect, the present disclosure provides a method performed by a user equipment (UE), the method including: the UE receiving, from an access network node, a portion of a model for generating decision, prediction, or output parameters using a wireless link between the UE and the access network node; if a wireless link failure occurs after the portion of the model has been received from the access network node but before the remaining portion of the model has been received by the UE, discarding the portion of the model; and, when the wireless link between the UE and the access network node is re-established, sending a request for the model; and receiving the model from the access network node. The model may be an artificial intelligence or machine learning (AI / ML) model.

[0047] In another aspect, the present disclosure provides a method performed by a user equipment (UE), the method including: the UE receiving, from an access network node, a portion of a model for generating decision, prediction, or output parameters using a wireless link between the UE and the access network node; if a wireless link failure occurs after the portion of the model is received from the access network node but before the remaining portion of the model is received by the UE, maintaining the portion of the model in a memory of the UE; re-establishing the wireless link between the UE and the access network node or establishing a wireless link between the UE and another access network node; if the wireless link is re-established between the UE and the access network node, sending a notification of the portion of the model stored in the UE to the access network node and receiving the remaining portion of the model from the access network node; and if the wireless link is established with the other access network node, sending a notification of the portion of the model stored in the UE to the other access network node and receiving the remaining portion of the model from the other access network node.

[0048] The model may be an artificial intelligence or machine learning (AI / ML) model.

[0049] If portions of the model are received at the UE in multiple data transfer units, each data transfer unit may contain a sub-portion of the model, and notification of the portion of the model stored at the UE may include notification of the identity of the last data transfer unit received at the UE.

[0050] The notification of the identity of the last data transfer unit received at the UE may include notification of the identity of a radio resource control (RRC) segment or notification of the identity of a packet data convergence protocol (PDCP) sequence number (SN). If the radio link is established with another access network, the method may further include sending to the other access network node at least one of an indication of the identity of the model or an indication of the identity of the access network node from which the UE received part of the model.

[0051] When the radio link is re-established with the access network node, the method may further include sending a notification of the identity of the model to the access network node.

[0052] The UE may be in a radio resource control (RRC) connected state when the UE receives portions of the model from the access network node, and the UE may maintain a context associated with the RRC connected state after a radio link failure occurs. In another aspect, the present disclosure provides a method performed by a user equipment (UE), the method including receiving, from a first access network node, a portion of a model for generating decision, prediction, or output parameters; performing a handover procedure for handover of the UE from the first access network node to a second access network node; maintaining the portion of the model in a memory of the UE during the handover procedure; sending a notification of the portion of the model stored in the UE to the second access network node; and receiving the remaining portion of the model from another access network node.

[0053] The method may further include receiving notification from the first access network node or the second access network node that the UE will receive the remaining portion of the model from the second access network node.

[0054] The model may be an artificial intelligence or machine learning (AI / ML) model. If portions of the model are received at the UE from the first access network node in multiple data transfer units, each data transfer unit may include a sub-portion of the model, and the notification of the portion of the model stored at the UE may include notification of the identity of the last data transfer unit received at the UE.

[0055] The notification of the identity of the last data transfer unit received at the UE may include notification of the identity of a radio resource control (RRC) segment or notification of the identity of a packet data convergence protocol (PDCP) sequence number (SN). In another aspect, the present disclosure provides a method implemented by an access network node, the method including the access network node transmitting, to user equipment (UE), a portion of a model for generating decision, prediction, or output parameters using a wireless link between the UE and the access network node, and if a wireless link failure occurs after the portion of the model is transmitted to the UE but before the remaining portion of the model is transmitted to the UE, re-establishing the wireless link between the UE and the access network node, receiving an indication from the UE of the portion of the model stored in the UE, determining the remaining portion of the model to be transmitted to the UE based on the indication of the portion of the model stored in the UE, and transmitting the remaining portion of the model to the UE.

[0056] The model may be an artificial intelligence or machine learning (AI / ML) model. If portions of the model are transmitted to the UE in multiple data transfer units, each data transfer unit may contain a sub-portion of the model, and notification of the portion of the model stored at the UE may include notification of the identity of the last data transfer unit received at the UE.

[0057] The notification of the identity of the last data transfer unit received at the UE may include notification of the identity of a radio resource control (RRC) segment or notification of the identity of a packet data convergence protocol (PDCP) sequence number (SN). The UE may be in a radio resource control (RRC) connected state when the access network node sends portions of the model to the UE, and the access network node may maintain the context associated with the RRC connected state after a radio link failure occurs. In another aspect, the present disclosure provides a method performed by a first access network node, the method including: sending to user equipment (UE) a portion of a model for generating decisions, predictions, or output parameters; sending a notification to the UE that a remaining portion of the model should be received from a second access network node; and performing a handover procedure for handover of the UE from the first access network node to the second access network node.

[0058] The model may be an artificial intelligence or machine learning (AI / ML) model. The method may further include transmitting the remaining portion of the model to the second access network node for transmission of the remaining portion of the model from the second access network node to the UE.

[0059] The method may further include sending a notification of the identity of the model to the second access network node.

[0060] In another aspect, the present disclosure provides a method performed by a second access network node, the method including: performing a handover procedure for handover of a UE from a first access network node to the second access network node; receiving from the UE or from the first access network node notification of a portion of a model for generating decision, prediction, or output parameters to be stored in the UE or notification of a remaining portion of the model to be transmitted to the UE; and transmitting the remaining portion of the model to the UE.

[0061] The model may be an artificial intelligence or machine learning (AI / ML) model.

[0062] The method may further include receiving a remaining portion of the model from the first access network node.

[0063] The method may further comprise receiving notification of the identity of the model from the first access network node.

[0064] In another aspect, the present disclosure provides a method performed by a user equipment (UE), the method including receiving a model for generating a determination, prediction, or output parameter; determining to activate the model for use in the UE; determining to activate the model for use in the UE; sending a notification to an access network node that the model has been activated for use in the UE; and sending a notification to the access network node that the model has been activated for use in the UE.

[0065] The model may be an artificial intelligence or machine learning (AI / ML) model. Determining to send a notification that the model has been activated for use at the UE may include determining to send a notification to an access network node when the model is received in a broadcast transmission.

[0066] In another aspect, the present disclosure provides a method performed by a user equipment (UE), the method including receiving a model for generating determination, prediction, or output parameters; receiving notification from an access network node that the model is to be activated for use in the UE; determining, based on the notification, to activate the model for use in the UE; and activating the model for use in the UE.

[0067] The model may be an artificial intelligence or machine learning (AI / ML) model. In another aspect, the present disclosure provides a method in a user equipment (UE), the method including receiving, from an access network node, notification of models for generating decision, prediction, or output parameters that are supported for use in a cell of the access network node, and sending to the access network node a notification that the UE is able to receive and use the models.

[0068] The method may further include receiving a request from the access network node for notification of whether the UE can receive and use the model, and the UE sending a notification of whether it can receive and use the model to the access network node after receiving the request.

[0069] The model may be an artificial intelligence or machine learning (AI / ML) model. The indication of whether the UE can receive and use the model may include an indication of at least one of a state of memory resources at the UE, a state of processing resources at the UE, or a state of power resources at the UE.

[0070] In another aspect, the present disclosure provides a method of an access network node, the method including sending to user equipment (UE) a notification of models for generating decision, prediction, or output parameters that are supported for use in a cell of the access network node; receiving from the UE a notification that the UE can receive or use the models; and determining based on the received notification whether the models should be sent to the UE or activated for use in the UE. The model may be an artificial intelligence or machine learning (AI / ML) model.

[0071] The indication of whether the UE can receive and use the model may include an indication of at least one of a state of memory resources at the UE, a state of processing resources at the UE, or a state of power resources at the UE.

[0072] In another aspect, the present disclosure provides an access network node, comprising: means for transmitting to a user equipment (UE) an indication of features for implementation in a cell of the access network node, the features being implemented using corresponding models for generating determination, prediction, or output parameters; and means for receiving from the UE a request for the models, wherein the transmitting means is configured to transmit to the UE model transmission information comprising an indication of one or more communications resources to be used by the UE to receive the models, or at least one of the models.

[0073] In another aspect, the present disclosure provides an access network node including: means for receiving, from a user equipment (UE), UE capability information indicating at least one of features supported by the UE, where the feature is implemented using a corresponding model for generating the determination, prediction, or output parameter, a model for generating the determination, prediction, or output parameter supported by the UE, or an indication of one of the models stored in the UE; means for determining, based on the UE capability information, a model to be transmitted to the UE or activated in the UE; and means for transmitting, to the UE, at least one of model transmission information including an indication of one or more communication resources to be used by the UE to receive the model, the model, or a UE request to activate the model.

[0074] In another aspect, the present disclosure provides a user equipment (UE) comprising: means for receiving, from an access network node, notification of features for implementation in a cell of the access network node, the features being implemented using a corresponding model for generating determination, prediction, or output parameters; means for determining to obtain the model; and means for transmitting a request for the model to the access network node, wherein the receiving means is configured to receive from the access network node model transmission information including an indication of one or more communication resources to be used by the UE to receive the model, or at least one of the models.

[0075] In another aspect, the present disclosure provides a user equipment (UE) including: means for transmitting, to an access network node, UE capability information indicating at least one of features supported by the UE, where the feature is implemented using a corresponding model for generating the determination, prediction, or output parameter, a model for generating the determination, prediction, or output parameter supported by the UE, or an indication of one of the models stored in the UE; and means for receiving from the access network node at least one of model transmission information including an indication of one or more communication resources to be used by the UE to receive the model, the model, or a UE request to activate the model.

[0076] In another aspect, the present disclosure provides a user equipment (UE), the UE being configured to receive, from an access network node, a portion of a model for generating decision, prediction, or output parameters using a wireless link between the UE and the access network node, and if a wireless link failure occurs after the portion of the model has been received from the access network node but before the remaining portion of the model has been received by the UE, to discard the portion of the model, and when the wireless link between the UE and the access network node is re-established, to send a request for the model and to receive the model from the access network node.

[0077] In another aspect, the present disclosure provides a user equipment (UE) configured for: receiving, from an access network node, a portion of a model for generating decision, prediction, or output parameters using a wireless link between the UE and the access network node; if a wireless link failure occurs after the portion of the model is received from the access network node but before the remaining portion of the model is received by the UE; maintaining the portion of the model in a memory of the UE; re-establishing the wireless link between the UE and the access network node or establishing a wireless link between the UE and another access network node; when the wireless link is re-established between the UE and the access network node, sending a notification of the portion of the model stored in the UE to the access network node and receiving the remaining portion of the model from the access network node; and when the wireless link is established with the other access network node, sending a notification of the portion of the model stored in the UE to the other access network node and receiving the remaining portion of the model from the other access network node.

[0078] In another aspect, the present disclosure provides a user equipment (UE) including: means for receiving, from a first access network node, a portion of a model for generating decision, prediction, or output parameters; means for performing a handover procedure for handover of the UE from the first access network node to a second access network node; means for maintaining the portion of the model in a memory of the UE during the handover procedure; means for sending a notification of the portion of the model stored in the UE to the second access network node; and means for receiving the remaining portion of the model from another access network node.

[0079] In another aspect, the present disclosure provides an access network node configured to transmit, to user equipment (UE), a portion of a model for generating decision, prediction, or output parameters using a wireless link between the UE and the access network node, and if a wireless link failure occurs after the portion of the model is transmitted to the UE but before the remaining portion of the model is transmitted to the UE, re-establish the wireless link between the UE and the access network node, receive from the UE an indication of the portion of the model stored in the UE, determine based on the indication of the portion of the model stored in the UE the remaining portion of the model to be transmitted to the UE, and transmit the remaining portion of the model to the UE.

[0080] In another aspect, the present disclosure provides a first access network node including: transmitting means configured for transmitting to a user equipment (UE) a portion of a model for generating a decision, prediction, or output parameter and for transmitting a notification to the UE that a remaining portion of the model should be received from a second access network node; and means for performing a handover procedure for handover of the UE from the first access network node to the second access network node.

[0081] In another aspect, the present disclosure provides a second access network node, the second access network node including: means for performing a handover procedure for handover of a UE from a first access network node to the second access network node; means for receiving from the UE or from the first access network node notification of a portion of a model for generating decision, prediction, or output parameters to be stored in the UE or notification of a remaining portion of the model to be transmitted to the UE; and means for transmitting the remaining portion of the model to the UE.

[0082] In another aspect, the present disclosure provides a user equipment (UE) including: means for receiving a model for generating a determination, prediction, or output parameter; means for determining to activate the model for use in the UE; means for activating the model for use in the UE; means for determining to send a notification to an access network node that the model has been activated for use in the UE; and means for sending a notification to the access network node that the model has been activated for use in the UE.

[0083] In another aspect, the present disclosure provides a user equipment (UE) including receiving means configured to receive a model for generating a determination, prediction, or output parameter and to receive notification from an access network node that the model should be activated for use in the UE; means for determining, based on the notification, to activate the model for use in the UE; and means for activating the model for use in the UE.

[0084] In another aspect, the present disclosure provides a user equipment (UE) including: means for receiving, from an access network node, notification of models for generating decision, prediction, or output parameters supported for use in a cell of the access network node; and means for sending, to the access network node, notification that the UE is able to receive and use the models.

[0085] In another aspect, the present disclosure provides an access network node including: means for sending to a user equipment (UE) a notification of models for generating decision, prediction, or output parameters that are supported for use in a cell of the access network node; means for receiving from the UE a notification that the UE can receive or use the models; and means for determining, based on the received notification, whether the models should be sent to the UE or activated for use in the UE. [Brief explanation of the drawings]

[0086] Embodiments of the present disclosure will now be 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”) telecommunications system. [Figure 2] FIG. 2 is a diagram illustrating a typical frame structure that may be used in the telecommunications system of FIG. [Figure 3] FIG. 3 is a schematic block diagram showing the main components of a DU 50 that may be used as part of a RAN base station 5 for the communication system 1 shown in FIG. [Figure 4] FIG. 4 is a schematic block diagram showing the main components of a CU 60 that may be used as part of a RAN base station 5 for the communications system 1 shown in FIG. [Figure 5] FIG. 5 illustrates a mobility procedure in which a handover occurs from a source base station to a target base station. [Figure 6]FIG. 6 illustrates a random access (RA) procedure that may be performed in the system of FIG. [Figure 7] FIG. 7 shows a schematic diagram of point-to-point and point-to-multipoint transmission. [Figure 8] Figure 8 shows a framework for an AI / ML model. [Figure 9] FIG. 9 illustrates a method for 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 illustrating an example of an AI / ML information update. [Figure 12] FIG. 12 illustrates an example of how the base station 5 may broadcast notification of supported AI / ML models. [Figure 13] FIG. 13 is a diagram illustrating an example in which an AI / ML model is transmitted from an AI / ML server to a UE via a base station. [Figure 14] FIG. 14 is a diagram illustrating an example in which an AI / ML model is transmitted from a CU of a distributed base station to a UE via a DU. [Figure 15] FIG. 15 illustrates an example in which a network is configured to notify one or more UEs of updates to an AI / ML model using paging. [Figure 16] FIG. 16 is a diagram illustrating an example of an AI / ML model function area. [Figure 17] FIG. 17 is a diagram illustrating how AI / ML model area information is received by a UE. [Figure 18] FIG. 18 is a diagram illustrating how the AI / ML model is transmitted by the base station to the UE 3. [Figure 19] FIG. 19 illustrates a variation of the example of FIG. 18, in which the AI / ML model is transmitted from the base station to the UE without providing a separate indication of the communication resources for transmitting the AI / ML model. [Figure 20]FIG. 20 illustrates a variation of the example of FIG. 18 in which the AI / ML model is transmitted from the AI / ML server to the UE. [Figure 21] FIG. 21 is a diagram showing an example in which a UE transmits UE AI / ML capability information to a base station 5 and receives an AI / ML model from the base station. [Figure 22] FIG. 22 illustrates an example in which the UE transmits UE AI / ML capability information to the base station, and the AI / ML model is transmitted from the base station to the UE without providing a separate notification of communication resources for AI / ML model transmission. [Figure 23] FIG. 23 is a diagram showing an example in which the UE 3 transmits UE AI / ML capability information to the base station 5, and an AI / ML model is transmitted from the AI / ML server to the UE. [Figure 24] FIG. 24 is a schematic block diagram illustrating the main components of a UE for the telecommunications system of FIG. [Figure 25] FIG. 25 is a schematic block diagram showing the main components of a base station for the telecommunications system of FIG. [Figure 26] FIG. 26 is a schematic block diagram showing the main components of a core network node or function for the telecommunications system of FIG. DETAILED DESCRIPTION OF THE INVENTION

[0087] overview An exemplary telecommunications system will now be described in general terms, by way of example only, with reference to FIGS.

[0088] FIG. 1 is a schematic diagram of a mobile (“cellular” or “wireless”) telecommunications system 1 to which exemplary embodiments of the present disclosure are applicable.

[0089] In network 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 ((R)AN) nodes 5 that operate according to one or more compatible radio access technologies (RATs). In the illustrated example, the (R)AN nodes 5 comprise NR / 5G base stations 5 or "gNBs" 5 that operate one or more associated cells 9. Communications via the base stations 5 are typically routed through a core network 7 (e.g., a 5G core network or an evolved packet core network (EPC)).

[0090] As one skilled in the art will appreciate, although three UEs 3 and one base station 5 are shown in FIG. 1 for illustrative purposes, the system, when implemented, will typically include other base stations 5 and UEs 3.

[0091] Each base station 5 controls, directly or indirectly via one or more other nodes (e.g., home base stations, repeaters, remote radio heads, distributed units, etc.), one or more associated cells 9. It will be appreciated that the base stations 5 may be configured to support 4G, 5G, 6G, and / or any other 3GPP or non-3GPP communication protocols.

[0092] The UEs 3 and their serving base stations 5 are connected via a suitable air interface (such as, for example, the so-called "Uu" interface). Neighboring base stations 5 may be connected to each other via a suitable inter-base station interface (such as the so-called "X2" interface, "Xn" interface, etc.).

[0093] The core network 7 includes several logical nodes (or "functions") for supporting communications in the telecommunications 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 (AMF) 10-1, one or more Session Management Functions (SMF) 10-2, and several other functions 10-n.

[0094] The base stations 5 are connected to core network nodes via appropriate interfaces (or "reference points"), such as the N2 reference point between the base stations 5 and the AMF 10-1 for communication of control signaling, and the N3 reference point between the base stations 5 and each UPF 11 for communication of user data. The UEs 3 are each connected to the AMF 10-1 via a logical non-access stratum (NAS) connection over the N1 reference point (similar to the S1 reference point in LTE). It will be appreciated that the N1 communications are transparently routed via the base stations 5.

[0095] The one or more UPFs 11 are connected to an external data network (eg, an IP network such as the Internet) via a reference point N6 for the communication of user data.

[0096] The AMF 10-1 performs mobility management related functions, maintains a NAS signaling connection with each UE 3, and manages UE registration. The AMF 10-1 is also responsible for managing paging. The SMF 10-2 provides session management functions (forming part of the MME function in LTE) and also combines some control plane functions (provided by the Serving Gateway and Packet Data Network Gateway in LTE). The SMF 10-2 also allocates IP addresses to each UE 3.

[0097] The base stations 5 of the communication system 1 are configured to operate at least one cell 9 on an associated TDD carrier operating in unpaired spectrum. It will be appreciated that the base stations 5 may also operate at least one cell 9 on an associated FDD carrier operating in paired spectrum.

[0098] The base station 5 is also configured to transmit control information and user data over several downlink (DL) physical channels and to transmit several physical signals, and the UE 3 is configured to receive control information and user data over several DL physical channels and to transmit several physical signals, where the DL physical channels correspond to resource elements (REs) carrying information originating from higher layers and the DL physical signals correspond to REs used by the physical layer and not carrying information originating from higher layers.

[0099] 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 capacity of the PDSCH on a time and frequency basis. The PDSCH can carry various data items, including, for example, user data, UE-specific higher-layer control messages mapped down 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 a Master Information Block (MIB) to the UE. The PBCH, in conjunction with the PDCCH, also supports time and frequency synchronization, which aids in cell acquisition, selection, and reselection. The UE 3 may receive a synchronization signal block (SSB), and the UE 3 may assume that the reception opportunities for the PBCH, primary synchronization signal (PSS), and secondary synchronization signal (SSS) are within consecutive symbols, forming an SS / PBCH block. The 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 periodicity of the SSB transmission may be signaled to the UE using any appropriate signaling (e.g., per serving cell using ssb-periodicityServingCell). The periodicity value of the SSB may be, for example, 20 ms or greater. For initial cell selection, the UE 3 may be configured to assume that the SS burst occurs with a periodicity of 2 frames.The UE 3 may also be provided with notification of which SSBs within the 5 ms duration are transmitted (eg, using ssb-PositionsInBurst).

[0100] DL physical signals may include, for example, reference signals (RS) and synchronization signals (SS). A reference signal (sometimes known as a pilot signal) is a signal having a predefined special waveform known to both the UE 3 and the base station 5. Reference signals 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).

[0101] Similarly, the UE 3 is configured to transmit control information and user data via several uplink (UL) physical channels corresponding to REs carrying information originated from higher layers and UL physical signals corresponding to REs used at the physical layer that do not carry information originated from higher layers, and the base station 5 is configured to receive control information and user data via several UL physical channels corresponding to REs carrying information originated from higher layers and UL physical signals corresponding to REs used at the physical layer that do not carry information originated from higher layers. The physical channels may include, for example, a PUSCH, a physical uplink control channel (PUCCH), and / or a physical random-access channel (PRACH). The UL physical signals may include, for example, a demodulation reference signal (DMRS) for UL control / data signals and / or a sounding reference signal (SRS) used for UL channel measurement.

[0102] When the UE 3 first establishes a radio resource control (RRC) connection with the base station 5 via a cell 9, the UE 3 registers with the appropriate core network node (e.g., AMF, MME). The UE 3 is in the so-called RRC connected state, and the associated UE context is maintained by the network. When the UE 3 is in the so-called RRC idle or RRC inactive state, the UE 3 selects a suitable cell for camping, so that the network knows the approximate location (not necessarily at cell level) of the UE 3.

[0103] The base station 5 may be a base station 5 that is split between one or more distributed units (DUs) 50 and a central unit (CU) 60, where the CUs 60 typically perform higher level functions and communication with the next-generation core, and the DUs 50 perform lower level functions and communication over the air interface with nearby UEs 3 (i.e., within the cell operated by the 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:

[0104] gNB Central Unit (gNB-CU): A logical node that hosts the Radio Resource Control (RRC), Service Data Adaptation Protocol (SDAP), and Packet Data Convergence Protocol (PDCP) layers of a gNB (or the RRC and PDCP layers of an en-gNB) and controls the operation of one or more gNB-DUs. The gNB-CU terminates the so-called F1 interface, which is connected to the gNB-DU.

[0105] gNB Distributed Unit (gNB-DU): A logical node that hosts the Radio Link Control (RLC), Medium Access Control (MAC), and Physical (PHY) layers of a gNB or en-gNB, and its operation is partially controlled by the gNB-CU. One gNB-DU supports one or more cells. One cell is supported by only one gNB-DU. The gNB-DU terminates the F1 interface that connects to the gNB-CU.

[0106] gNB-CU-Control Plane (gNB-CU-CP): A logical node that hosts the control plane portion of the RRC and PDCP protocols of the gNB-CU for the 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.

[0107] gNB-CU-User Plane (gNB-CU-UP): A logical node that hosts the user plane portion of the PDCP protocol of the gNB-CU for the en-gNB, and the user plane portions of the PDCP protocol and SDAP protocol of the gNB-CU for the gNB. The gNB-CU-UP terminates the E1 interface connected to the gNB-CU-CP and the F1-U (F1 user plane) interface connected to the gNB-DU.

[0108] It will be appreciated that where a distributed base station or similar control plane-user plane (CP-UP) split is employed, the control plane entity and user plane entity may each include associated transceiver circuitry, antennas, network interfaces, controllers, memory, operating systems, and communication control modules. When the base station 5 comprises a distributed base station, the network interface also includes an E1 interface and an F1 interface (F1-C for the control plane and F1-U for the user plane) for communicating signals between the respective functions of the distributed base station.

[0109] Frame structure 2, which illustrates a typical frame structure that may be used in communication system 1, base stations 5 and UEs 3 of communication system 1 communicate with each other in the time domain using resources organized into frames of length 10 ms. Each frame comprises 10 equally sized subframes of length 1 ms. Each subframe is divided into one or more slots of equal length, each comprising 14 Orthogonal Frequency-Division Multiplexing (OFDM) symbols.

[0110] As can be seen in FIG. 2, communication system 1 supports multiple different numerologies (subcarrier spacing (SCS), slot length, and thus OFDM symbol length). Specifically, each numerology is identified by a parameter μ, where μ=0 represents 15 kHz (corresponding to the LTE SCS). Now, the 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]

[0111] (R)AN Node DU Figure 3 is a schematic block diagram illustrating the main components of a DU 50 that may be used as part of an (R)AN node 5 for the communications system 1 shown in Figure 1. As shown, the DU 50 has transceiver circuitry 451 for transmitting signals to and receiving signals from a communications device (such as a UE 3) via a radio unit (RU) and associated DU-RU interface 453, and for transmitting signals to and receiving signals from a CU 60 of the (R)AN node 5 via a CU interface 454 (e.g., comprising an F1 interface that may be divided into F1-U and F1-C interfaces for user plane and control plane signaling, respectively).

[0112] The DU 50 has a controller 457 for controlling the operation of the DU 50. The controller 457 is associated with a memory 459. Software may be pre-installed in the memory 459 and / or may be downloaded, for example, via the communication system 1 or from a removable data storage device (RMD). The controller 457 is configured, in this example, to control the overall operation of the DU 50 by program instructions or software instructions stored in the memory 459.

[0113] As shown, these software instructions include, among other things, an operating system 461, a communication control module 463, an F1 module 465, a DU-RU module 468, a DU management module 472, a UE profile management module 473, and a mobility module 475.

[0114] The communication control module 463 is operable to control communications between the DU 50 and one or more RUs (and thus between the DU 50 and the UE 3), and between the DU 50 and the CU 60. The communication control module 463 is configured to generally control the reception of signals corresponding to uplink communications from the UE 3, and to handle the transmission of downlink communications to the UE 3.

[0115] The F1 module 465 is responsible for appropriate processing of signals received from or transmitted to the CU 60 via one or more CU (e.g., F1) interfaces 454. These signals may be separated into user plane signals received from or transmitted to the CU-UP portion of the CU 60 via the F1-U interface and control plane signals received from or transmitted to the CU-CP portion of the CU 60 via the F1-C interface.

[0116] The DU-RU module 468 is responsible for the proper processing of signals received from or transmitted to the RUs via one or more RU (eg, DU-RU) interfaces 453 .

[0117] The DU management module 472 is responsible for managing the overall operation of the DU 50 and the overall performance of tasks required by the DU 50. These tasks include, among other things, generating and transmitting appropriate messages using an appropriate signaling application protocol depending on the functional division between the RU, DU 50, and CU 60, such as interpreting received MAC signaling and generating MAC signaling for transmission. The DU management module 472 may, where appropriate, control the overall operation of the DU 50 according to any of the methods described below.

[0118] The UE profile management module 473 is responsible for performing the following UE profile-related functions: (where applicable): receiving and storing a UE profile or associated assistance / preference information from the UE 3 or from elsewhere in the network; determining an appropriate mobility-specific configuration based on the UE profile / assistance / preference information for implementation in the UE 3 and / or the (R)AN node 5 (where applicable); and / or (where applicable) providing configuration information for appropriately configuring the UE in a mobility-based configuration. The UE profile management module 473 may also store, for example, previous mobility information for the UE 3 (e.g., previous movements of the UE 3 between different communication cells of the network). It will be understood that, depending on the implementation, the DU 50 may not implement at least some of these features.

[0119] The mobility module 475 is responsible for controlling mobility procedures for one or more UEs 3. For example, the mobility module 475 may be configured to perform one or more measurements of UE 3 mobility or to select candidate cells for handover.

[0120] CU Figure 4 is a schematic block diagram illustrating the main components of a CU 60 of an (R)AN node for the communication system 1 shown in Figure 1. As shown, the CU 60 has transceiver circuitry 551 for transmitting signals to and receiving signals from the DUs 50 via one or more DU interfaces 554 (e.g., comprising an F1 interface that may 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 functions of the core network 7 via one or more CU network interfaces 555 (e.g., including N2 and N3 interfaces, etc.).

[0121] The CU 60 has a controller 557 for controlling the operation of the CU 60. The controller 557 is associated with a memory 559. Software may be pre-installed in the memory 559 and / or may be downloaded, for example, via the communication system 1 or from a removable data storage device (RMD). The controller 557 is configured, in this example, to control the overall operation of the CU 60 by means of program or software instructions stored in the memory 559.

[0122] As shown, these software instructions include, among other things, an operating system 561, a communication control module 563, an F1 module 565, an E1 module 566, an N2 module 568, an N3 module 569, a CU-UP management module 571, a CU-CP management module 572, a UE profile management module 573, and a mobility module 575. The functionality of mobility module 575 is as previously described with reference to FIG.

[0123] The communication control module 563 is operable to control communications between the CU 60 and one or more DUs 50 (and thus between the CU 60 and the UE 3), and between the CU 60 and the core network 7. The communication control module 563 is configured to generally control the reception of signals corresponding to uplink communications from the UE 3, and to control the transmission of downlink communications.

[0124] The F1 module 565 is responsible for the appropriate 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 at or transmitted by the CU-UP portion of the CU 60 via the F1-U interface, and control plane signals received at or transmitted by the CU-CP portion of the CU 60 via the F1-C interface.

[0125] The E1 module 566 is responsible for the proper processing of signals transmitted between the CU-UP portion of the CU 60 and the CU-CP portion of the CU 60 via a corresponding internal CU interface (eg, E1).

[0126] The N2 module 568 is responsible for the appropriate processing of signals received from or transmitted to the AMF 8-1 via one or more corresponding CU interfaces 555 (eg, N2). The N3 module 569 is responsible for the appropriate processing of signals received from or transmitted to one or more core network user plane functions via one or more corresponding CU interfaces 555 (e.g., N3).

[0127] The CU-UP management module 571 is responsible for managing the overall operation of the CU-UP portion of the CU 60 and the overall performance of tasks required for the CU-UP.

[0128] The CU-CP management module 572 is responsible for managing the overall operation of the CU-CP portion of the CU 60 and the overall performance of tasks required of the CU-CP, including, among other things, the generation and transmission of appropriate messages using the appropriate signaling application protocol depending on the functional division between the RU, the DU 50, and the CU 60, such as interpreting received RRC signaling and generating RRC signaling for transmission.

[0129] The UE profile management module 573 is responsible for performing the following functions related to UE (mobility) profiles: (where applicable): receiving and storing a UE profile or associated assistance / preference information from the UE 3 or from elsewhere in the network; determining an appropriate mobility-specific configuration based on the UE profile / assistance / preference information for implementation in the UE 3 and / or the (R)AN node 5; and / or providing configuration information for appropriately configuring the UE in a mobility-based configuration. The UE profile management module 573 may also store previous mobility information for the UE 3 (e.g., previous movements of the UE 3 between different communication cells of the network). It will be understood that, depending on the implementation, the CU 60 may not implement at least some of these features.

[0130] System Information and SIB It will be appreciated that transmissions in the cell 9 of the base station 5 may include one or more broadcast transmissions, one or more unicast transmissions for reception by a UE 3, and / or one or more multicast transmissions for reception by a group of UEs 3. The system information (SI) transmitted in the cell may include a "minimum SI" (MSI) and an "other SI" (OSI). The OSI may be broadcast on demand, for example, using the downlink shared channel (DL-SCH). The OSI may be broadcast upon request from a UE 3 in a radio resource control (RRC) idle or RRC inactive state. The OSI may also be requested by a UE 3 in an RRC connected state, for example, via one or more dedicated RRC transmissions.

[0131] The SI may include information to enable (e.g., configure) the UE 3 to complete a cell selection, to enable the UE 3 to complete a cell reselection procedure, or to enable the UE 3 to receive one or more paging messages transmitted within the cell. The SI may be broadcast using a Master Information Block (MIB) and one or more System Information Blocks (SIBs).

[0132] The MSI comprises a MIB and a system information block 1 (SIB1). The MIB includes information for the UE 3 to use to receive SIB1, such as the subcarrier spacing of SIB1. The MIB provides information corresponding to the Control Resource Set (CORESET) and the search space. SIB1 may be referred to as the "remaining MSI" (RMSI). SIB1 may be transmitted in a dedicated RRC message, and other SIBs (such as SIB2 to SIB9) may be transmitted using one or more other appropriate RRC transmissions (such as another dedicated RRC message). The MIB and SIB1 may provide the UE 3 with notification of scheduling information for receiving and decoding other SIBs, such as SIB2 to SIB9, and may provide information for the UE 3 to use to receive one or more paging messages. The OSI may include, for example, SIB2 to SIB9 transmitted using the DL-SCH in an SI message. Mapping of SIB2 to SIB9 to their corresponding SI messages may be provided to the UE 3 by the base station 5. The MIB and SIB1 to SIB9 are described in more detail, for example, in 3GPP 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 for 4G (LTE). SIB6 and SIB7 provide information for the earthquake and tsunami warning system (ETWS). SIB8 provides information for commercial mobile alert service (CMAS) notifications, for example to provide warning text messages to UE3. SIB9 contains information on harmonized universal time (UTC), global positioning system (GPS) time (e.g., for GPS initialization), and local time.

[0133] The SIBs may be broadcast periodically (e.g., according to a predetermined periodic pattern) or alternatively, may be provided "on-demand," e.g., in response to a request from UE 3. For example, MIB may be transmitted with a period of 80 ms and repetition occurring within 80 ms, while SIB1 may be transmitted with a period of 160 ms and a variable transmission repetition period (e.g., 20 ms) within 160 ms. SIB1 may be used to indicate to UE 3 which SIBs are transmitted periodically and which SIBs are available on-demand in response to a request from UE 3. UE 3 may be configured to request on-demand SIBs using message 1 (MSG1), which may be referred to as an MSG1-based on-demand SI request, or message 3 (MSG3), which may be referred to as an MSG3-based on-demand SI request.

[0134] A physical broadcast channel (PBCH) can be used to broadcast the MIB. Base station 5 may transmit the PBCH together with a synchronization signal (SS) (e.g., a primary synchronization signal (PSS) and a secondary synchronization signal (SSS)) in an SS / PBCH block. The SS / PBCH block comprises four orthogonal frequency-division multiplexed (OFDM) symbols that are mapped to the PSS, SSS, and PBCH associated with a demodulation reference signal (DM-RS). In the frequency domain, the SS / PBCH block includes 240 consecutive subcarriers. When UE 3 is in an RRC connected state, base station 5 may provide UE 3 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 be transmitted similarly, for example, using the PDSCH. When one or more beamformed transmissions are transmitted in a cell served by base station 5, only some of the SIs (e.g., some of the SIBs) may be transmitted using a particular beam or using a particular transmission / reception point (TRP).

[0135] UE Mobility Figure 5 shows an overview of a mobility procedure that may be performed in a communications system 1 of the type illustrated in Figure 1. In this example, a handover of a UE 3 from a source base station 5 to a target base station 5 is performed.

[0136] In optional step S501, the UE 3 performs measurements. The measurements may be measurements of signals transmitted by the source (R)AN node 5 or measurements of signals transmitted by the target (R)AN node 5. The measurements may be measurements of signal strength that may be used as part of the determination that the UE 3 should be handed over from the source (R)AN node 5 to the target (R)AN node 5. In optional step S502, the UE 3 sends a measurement report to the source (R)AN node 5 providing notification of the results of the measurements. The measurement report may be sent from the UE 3 to the source base station 5 in an RRC message. In this example, the source (R)AN node 5 uses information provided in the measurement report to determine that the UE 3 should be handed over to the target (R)AN node 5. However, it will be appreciated that the determination that a handover to the target (R)AN node 5 should be performed may alternatively (or additionally) be based on measurements performed in the source (R)AN node 5 or the target (R)AN node 5. Alternatively, the determination that a handover of the UE 3 should be performed may be based on factors other than signal measurements, such as the level of congestion in a cell operated by the source (R)AN node 5, or inferences (e.g., determinations or predictions) generated using AI / ML models.

[0137] In step S503, the source (R)AN node 5 sends a handover request to the target (R)AN node 5, requesting handover of the UE 3 from the source (R)AN node 5 to the target (R)AN node 5. The handover request may include, for example, an indication of the identity of the source (R)AN node 5, a cause value for the handover, the identity of the target cell, UE 3 context information (e.g., the maximum bit rate of the UE 3 or the security capabilities of the UE 3), and UE history information. If the handover was triggered by a measurement report received by the source (R)AN node 5 in step S502, the cause value may indicate, for example, that the handover is desirable for radio reasons. Alternatively, if the handover was triggered to reduce the load at the source (R)AN node 5, the cause value may indicate that the handover is for reducing the load at the serving cell. The handover request message may also include an indication of the AMF 10-1 serving the UE 3.

[0138] In step S504, the target (R)AN node 5 sends an acknowledgement of the handover request (which may be referred to as a "Handover Request Acknowledgement" message). The Handover Request Acknowledgement message includes notification of handover configuration information for the handover to be forwarded to the UE 3. The Handover Request Acknowledgement message may also include configuration information that enables the source (R)AN node 5 to begin forwarding user plane data for the UE 3 to the target (R)AN node 5.

[0139] The transmissions of 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 (hence, the handover procedure in this example may be referred to as an Xn-based handover procedure). Steps S501 to S504 may be referred to as a "handover preparation phase."

[0140] In step S505, the source (R)AN node 5 sends handover configuration information to the UE 3. The configuration information for the handover may be, for example, an RRC configuration sent in an RRC configuration message or an RRC reconfiguration message. In step S506, the UE 3 applies the received configuration 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".

[0141] Following the handover execution phase, the UE 3 is operable to send uplink transmissions (e.g., uplink data) to the target (R)AN node 5 and receive downlink transmissions (e.g., downlink data) from the target (R)AN node 5.

[0142] It will be understood that the mobility method and handover procedure for the UE 3 are not limited to the example shown in Figure 5. For example, the UE 3 may be configured to perform a conditional handover (CHO) in which the UE 3 determines whether to handover the UE 3 to a candidate cell based on one or more execution conditions. It will also be understood that a handover in which the DU 50 changes but the CU 60 remains the same (inter-DU intra-CU handover), a handover in which both the DU 50 and the CU 60 change (inter-DU inter-CU handover), or a handover between two cells served by the same DU 50 can be performed.

[0143] Random Access Figure 6 shows a random access (RA) procedure that may be performed in the system of Figure 1. The RA procedure may be used, for example, for initial access by a UE 3 in RRC idle mode or for transitioning from RRC inactive mode to RRC connected mode. The RA procedure may also be used during handover of a UE 3 from a source base station to a target base station (e.g., such as the handover procedure described above with reference to Figure 5) for initial access to the target base station 5.

[0144] In step S601, the UE 3 transmits a random access preamble to the base station 5. In this example, the UE 3 selects a random access preamble to transmit from a group of random access preambles shared with other UEs 3. The transmission in step S601 may be referred to as message 1 (MSG1), and is transmitted using the PRACH.

[0145] In step S602, the base station 5 transmits a random access response to the UE 3. The transmission of step S602 may be referred to as message 2 (MSG2). The random access response indicates time and / or frequency resources (e.g., resource blocks and / or symbols) that the UE 3 should use to send subsequent transmissions to the base station 5. The random access response may also include further information for the UE 3 to use for communication with the base station 5, such as a timing advance (TA) value.

[0146] In step S603, the UE 3 sends a transmission to the base station 5 using the notified time and / or frequency resources. The transmission of step S603 may be referred to as message 3 (MSG3). The transmission of step S603 may be a layer 2 (L2) or layer 3 (L3) message. The transmission of step S603 may include, for example, an RRC setup request, an RRC resume request, an RRC re-establishment request, or an RRC reconfiguration complete message.

[0147] If two UEs 3 select and transmit the same random access preamble in step S601 and receive and decode MSG2 transmitted by the base station 5 in step S602, the two UEs may transmit MSG3 using the same time and / or frequency resources. This situation may be referred to as a "contention" or "collision." To resolve the contention, the base station 5 transmits a content resolution message to the UE 3 in step S604. The transmission in step S604 may be referred to as message 4 (MSG4). MSG4 indicates to the UE 3 whether the base station successfully received and decoded the MSG3 transmitted by the UE 3 in step S603. If the base station 5 decodes the MSG3 transmitted by another UE 3 that conflicts with the UE 3, or if interference occurs between the MSG3 transmitted by the two UEs 3, the MSG3 transmitted in step S603 may not be successfully received or decoded by the base station 5. If MSG3 sent by UE3 is not 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 sends another MSG1 to base station 5 (e.g., after selecting a different random access preamble).

[0148] The procedure shown in Figure 6 is an example of a contention-based RA procedure in which UE 3 selects a random access preamble from a group of preambles that other UEs 3 can also use (thus, a conflict may occur if two UEs 3 select the same random access preamble). Alternatively, base station 5 may send a random access preamble assignment to UE 3 before UE 3 sends MSG1 to base station 5, in which case the RA procedure is contention-free (and conflict resolution in step S604 may not be performed). The random access preamble assignment may be sent to UE 3 using an RRC message or layer 1 (L1) signaling (e.g., using DCI carried by the PDCCH). In the method shown in Figure 5, a random access preamble assignment for communicating with target base station 5 may be sent to UE 3 in step S505.

[0149] MSG1 and / or MSG3 may be used by UE3 to request on-demand SI from base station 5.

[0150] Broadcast and Multicast A base station 5 may transmit a broadcast intended for reception by any UE 3 within the cell of the base station 5, or may transmit a transmission (such as a point to point (PTP) transmission) intended for reception by a specific UE 3. A base station 5 may also transmit a transmission (such as a point to multiple (PTM) transmission) intended for reception by a specific group of UEs 3. A transmission intended for reception by a single UE 3 may be referred to as a unicast transmission, and a transmission intended for reception by a group of UEs 3 may be referred to as a multicast transmission.

[0151] A multicast service may include a PTP leg between a base station 5 and a single UE 3, and PTM legs between a base station 5 and multiple UEs 3. PTP and PTM transmissions are shown schematically in Figure 7. Although the UEs 3 are shown separately in Figure 7, it will be understood that the UEs 3 may receive both the PTP and PTM portions of a multicast. A PTP may also be described as a PTP "leg" or "portion" of a multicast transmission. Similarly, a PTM may also be described as a PTM "leg" or "portion" of a multicast transmission.

[0152] The PTM leg has an MBS radio bearer (MRB) with a corresponding MRB configuration. Each MRB may have an associated identifier (e.g., an MRB identity) that can be used to identify the MRB. The MRB identity may be included in any appropriate transmission for MRB configuration. The multicast service may be suspended (the process by which an MRB is released) or reactivated based on multicast data activity (or inactivity). The configuration of one or more MRBs may be provided to the UE 3 and / or the base station using any appropriate radio link control (RLC) configuration signaling (e.g., in an RLC bearer configuration message).

[0153] The base station 5 may provide the multicast MRB configuration to the UE 3 via dedicated signaling. The multicast MRB may be configured in DL-only RLC unacknowledge mode (RLC-UM) where no acknowledge / negative-acknowledge (ACK / NACK) feedback is sent, or the MRB may have a bidirectional RLC-UM configuration for PTP transmission. The multicast MRB configuration may include an RLC-acknowledge mode (RLC-AM) configuration for transmitting ACK / NACK feedback. The multicast MRB configuration may include an RLC-unacknowledge mode (RLC-UM) configuration for not transmitting ACK / NACK feedback.

[0154] The multicast MRB configuration may include an RLC-AM entity for PTP transmission. The multicast MRB configuration may include a DL-only RLC-UM entity for PTM transmission. A multicast MRB configuration may include two RLC-UM entities: one RLC-UM entity may be a DL-only RLC-UM entity for PTP transmissions, and the other RLC-UM entity may be a DL-only RLC-UM entity for PTM transmissions.

[0155] A multicast MRB configuration may include three RLC-UM entities, one of the RLC-UM entities being a DL-only RLC-UM entity, one of the RLC-UM entities being a UL RLC-UM entity for PTP transmission, and the other RLC-UM entity being a DL-only RLC-UM entity for PTM transmission.

[0156] A multicast MRB configuration may include two RLC entities, one RLC entity being an RLC-AM entity for PTP transmission and the other RLC entity being a DL-only RLC-UM entity for PTM transmission.

[0157] Logical Channels and Logical Channel Priorities A logical channel (LCH) may be a control channel for transmitting control and / or configuration information (control plane information) or a channel used for transmitting user data (user plane information). Logical channels that may be used in the system shown in FIG. 1 include a broadcast control channel (BCCH), a paging control channel (PCCH), a common control channel (CCCH) used by UE 3 during initial access, a dedicated control channel (DCCH), and a dedicated traffic channel (DTCH). The system of FIG. 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 transmission using a transport channel (e.g., based on the priority of each logical channel, as described below). For example, the BCCH may be mapped to the BCH or DLSCH, and the PCCH may be mapped to the PCH. A transport channel is mapped to a corresponding physical channel (e.g., a PDCCH, PDSCH, or PBCH for downlink transmission; or a PUSCH, PUCCH, or PUSCH for uplink transmission).

[0158] A logical channel may be identified using a corresponding logical channel ID (LCID), and a set of logical channels may be grouped into a logical channel group (LCG), which may be identified using a corresponding index (e.g., an index between 0 and 7).

[0159] Logical channels may be assigned priorities (e.g., transmission priority) by the network. For example, logical channels used for part of a handover procedure may be assigned a relatively high priority for transmission, since transmission delays in the handover procedure increase the likelihood of handover failure. The base station 5 may determine to preferentially include data (or other information) corresponding to a high-priority logical channel in a medium access control (MAC) protocol data unit (PDU) for transmission to the UE 3, rather than including data or other information corresponding to a lower-priority logical channel. The base station 5 may also control the scheduling of uplink transmissions by the UE 3 based on the logical channel priorities.

[0160] A prioritized bit rate (PBR) may be defined for a logical channel. The prioritized bit rate may be set by the base station 5. The prioritized bit rate is the bit rate configured to be used for 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 beneficially helps to avoid situations where only the highest priority logical channel is transmitted.

[0161] Artificial Intelligence(AI) / Machine Learning(ML) Figure 8 shows a framework for AI / ML models and how the various entities of the framework can interact with each other.

[0162] The entities include a data collection function 41, a model training function 43, a model inference function 45, and actors 47. The data collection function 41 provides input data (training data) to the model training function 43 and the model inference function 45. The collected data may be, for example, data related to mobility (e.g., handover of the UE 3 or the location of the UE 3). For example, the data may be acquired by a base station 5 (e.g., by receiving measurement reports from the UE 3 or by receiving data from another base station 5 or core network node / function) and transmitted to another base station 5, which generates the 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 may perform ML model training, validation, and testing and generate model performance metrics as part of the model testing procedure. The model reasoning function 45 provides an AI / ML model inference output (e.g., a prediction or a decision), and the actor 47 is a function or node that receives the output from the model reasoning function 45 and triggers or performs a corresponding action (e.g., the base station 5 increasing / decreasing its transmit power or initiating a handover procedure for the UE 3). The AI / ML model inference output may, for example, be a prediction of the mobility of the UE 3 (e.g., an expected path, route, or trajectory, inter-cell or inter-beam mobility, or an expected handover), or may be one or more parameters for use in encoding or decoding transmissions between the base station 5 and the UE 3. The functions shown in FIG. 8 may be co-located at a single node of the communications network (e.g., at a base station 5 or core network node / function) or distributed among multiple network nodes (e.g., multiple base stations 5).

[0163] Terms referred to by 3GPP in the context of this framework include: AI / ML training: An online or offline process for training an AI / ML model. AI / ML validation: A method for evaluating the quality (e.g., predictive accuracy) of an AI / ML model using a dataset different from 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 dataset different from the dataset used for training and validation. AI / ML Data Collection: A method for collecting data by a network node, management entity, and / or UE3 to train an AI / ML model, for data analysis (e.g., model performance monitoring), and / or to generate inferences using an AI / ML model. Model monitoring: A method for monitoring the inference performance (e.g., prediction accuracy) of AI / ML models. Training data: Data used as input to the AI / ML model training function. Supervised learning: How to train AI / ML models using labeled data. Unsupervised learning: How to train AI / ML models using unlabeled data. Semi-supervised learning: A method of training AI / ML models using both labeled and unlabeled data. Inference data: Data to input into an AI / ML model inference function to generate inferences. Model Deployment / Update: How to deploy an AI / ML model to a model inference facility (e.g., send it to a network node) or how to distribute an updated model to a model inference facility.

[0164] The data collection 41 may be performed at various nodes of the communication network (eg, at one or more base stations 5 or at the UE 3).

[0165] FIG. 9 illustrates a method for training an AI / ML model and monitoring the performance of the AI / ML model. As shown in FIG. 9, stored data / features may first be extracted in a data extraction step. In a data validation step, a decision is made (e.g., based on the extracted data) to proceed or retain training of the AI / ML model. In a data preparation stage, data is prepared for use in training the AI / ML model. For example, the data may be cleaned (e.g., filtered), transformed, or modified in any other suitable manner. The data may also be split into training data, validation data, and test data sets in the data preparation stage.

[0166] In the model training step, the AI / ML model is trained (or retrained) using the training data prepared in the data preparation step. It will be appreciated that any suitable training method can be used to train the AI / ML model (e.g., methods including supervised learning or unsupervised learning). In the model evaluation step, the AI / ML model is evaluated (e.g., the predictive accuracy of the AI / ML model is evaluated) using a test dataset (which may be generated in the data preparation step). In the model validation step, a determination is made (e.g., based on the results of the model evaluation step) as to whether the AI / ML model is suitable for deployment in a communications network.

[0167] In the model serving step, the AI / ML model is deployed for use in the communications network 1. AI / ML model deployment may include compiling the trained AI / ML model, packaging the model into an executable format, and distributing the AI / ML model to target devices. For example, the AI / ML model may be transmitted to the base station 5 and / or the UE 3 for use by the base station and / or the UE to generate predictions or decisions using the AI / ML model as part of the prediction serving step, as shown in the figure. In the performance monitoring step, the performance of the deployed AI / ML model is monitored. The predictive performance of the AI / ML model may be monitored by comparing predictions generated using the model to one or more measurements. For example, if the AI / ML model is used to predict the location of the UE 3, the predictive accuracy of the AI / ML model may be evaluated using measurements of the actual location of the UE 3. Alternatively, if the AI / ML model is used to determine parameters to use in encoding and decoding data transmitted between the base station 5 and the UE 3, 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 (e.g., because the predictive accuracy of the AI / ML model falls below an acceptable threshold accuracy, or because the performance of a method using inferences from the AI / ML model falls below an acceptable threshold performance), and the method returns to the data extraction step.

[0168] As described above with reference to FIG. 8, each step of the method of FIG. 9 may be performed at a single node of the communication system 1, or alternatively, the steps of the method may be distributed among multiple different nodes.

[0169] As discussed above with reference to Figures 8 and 9, information collected by nodes / functions in a communications network can be used as training data for an AI / ML model and can be used as inference data for use in generating one or more model inferences using the AI / ML model. Information used as training data and / or to generate one or more model inferences may be referred to as "AI / ML information." Methods for requesting and transmitting AI / ML information are now described.

[0170] 10 is a diagram showing an example of an AI / ML information request and an 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 (such as information about the actual mobility of the UE 3) from the second base station 5-2.

[0171] After receiving the AI / ML information request in step S1501, the second base station 5-2 transmits an AI / ML information response including the AI / ML information to the first base station 5-1. In addition, the second base station 5-2 may start periodically reporting the 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 a corresponding AI / ML information reporting configuration indicated by the AI / ML information request (including, for example, the periodicity of the reporting, the number of reports, or the duration / period of the reporting). The AI / ML information request may include an information element (IE) indicating that the second base station 5-2 starts or stops periodically reporting the 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 the AI / ML information from the second base station 5-2, rather than a periodic report.

[0172] If the second base station 5-2 is unable to transmit the requested AI / ML information to the first base station 5-1 (e.g., because the requested information is unavailable at the second base station 5-2), the second base station 5-2 may transmit a corresponding notification, e.g., an AI / ML information failure message, to the first base station 5-1 that the second base station 5-2 is unable to provide the requested information. The AI / ML information failure message may include an indication of the reason (e.g., a cause value) why the second base station 5-2 is unable to provide the requested AI / ML information.

[0173] Upon receiving the AI / ML information, the first base station 5-1 may use the AI / ML information to train (or update) a corresponding AI / ML model (e.g., with respect to the mobility of the UE 3) or to generate a prediction (e.g., a prediction of the mobility of the UE 3). Alternatively, the first base station 5-1 may forward the AI / ML information to another network node for use with an AI / ML model at that network node.

[0174] In the example shown in FIG. 10, the AI / ML information response may include the requested AI / ML information, but 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 (e.g., an acknowledgment of the AI / ML information request). FIG. 11 is a diagram showing an example of an AI / ML information update. In step S1601, the second base station 5-2 determines to transmit an AI / ML information update to the first base station 5-1. For example, the second base station 5-2 may determine to transmit an AI / ML information update to the first base station 5-2 based on the reporting periodicity received by the second base station 5-2 in step S1501 of FIG. 10, or may determine to transmit an AI / ML information update based on a change in the AI / ML information stored in the second base station 5-2 (or based on new AI / ML information obtained by 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 an AI / ML information update.

[0175] Distributed AI / ML Architecture While a network may include a primary node / function that hosts the AI / ML model and generates the AI / ML model inferences, alternatively, the AI / ML model may be distributed among various nodes within the network. For example, multiple base stations 5 may host the AI / ML model and generate inferences. While this may increase the processing required at some network nodes, distributing the AI / ML model among network nodes reduces the number of inferences sent between nodes.

[0176] If an AI / ML model (or multiple AI / ML models—not necessarily the same model used at each node) is provided to multiple base stations 5, feedback information may still be provided to each base station that generates inferences using the AI / ML model (e.g., to verify the accuracy of the model, as described above).

[0177] AI / ML configuration information Configuration information for AI / ML models (sometimes referred to as "AI / ML configuration information") may be exchanged between nodes in a communication network. For example, a core network node may send the AI / ML configuration information to a base station 5 that hosts the AI / ML model. The AI / ML configuration information may include a list of supported use cases for the AI / ML model (although the AI / ML model does not necessarily have to be for predicting UE mobility). Supported use cases may be, for example, energy saving, traffic steering, anomaly detection, quality of experience (QoE) optimization, mobility robustness optimization (MRO), RAN slice service level agreement (SLA) guarantees, massive 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. The AI / ML configuration information may include an indication of the specific AI / ML model to use for a particular use case. The AI / ML configuration information may also include an indication of whether feedback is required (e.g., from another network node). The feedback may include, for example, communication performance feedback (eg, informing of communication performance for communication between the UE 3 and the base station 5).

[0178] If multiple AI / ML models are stored in the UE 3 (or base station 5, or other network node), the UE 3 may receive notification of which AI / ML model to use. The UE 3 may receive notification (e.g., from the base station 5) that use of a particular model should be activated or deactivated (e.g., responsive to a determination in the performance monitoring step of FIG. 9, a particular AI / ML model may be deactivated if prediction accuracy falls below an acceptable accuracy threshold). The UE 3 may be equipped with multiple AI / ML models, each model intended for use in a particular scenario or configuration.

[0179] Single and double sided models The AI / ML model may be hosted (stored to generate inferences) at both the base station 5 and the UE 3, or may be hosted only at the base station 5, or may be hosted only at the UE 3. When the AI / ML model is used only at the UE 3, the AI / ML model may be referred to as a "single-sided" model. For example, the UE 3 may host an AI / ML model for generating time (e.g., time resources) for communication using a particular beam transmitted by the base station 5. However, it should be understood that even if the model is a single-sided model, the model does not necessarily need to be trained at the UE 3. For example, the model may be trained at the base station 5 or another node in the network (e.g., a core network node / function) and then transmitted to the UE 3 for use by the UE 3. In other words, the AI / ML model may be trained at another network node and then transferred / deployed to the UE 3.

[0180] Alternatively, the AI / ML model may be a “two-sided” model in which the AI / ML model is hosted on the UE 3 and a corresponding AI / ML model is hosted on the base station 5 (however, the models do not necessarily have to be hosted on the UE 3 and the base station 5; any other suitable two network nodes may be used instead). The AI / ML model hosted on the UE 3 and the AI / ML model hosted on the base station 5 may be the same AI / ML model (but are not necessarily the same model). The UE 3 may use the AI / ML model to generate a first inference, and the base station 5 may 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 the UE 3 to the base station 5, and the second inference may be an inference of parameters used to decode or decompress the data at the base station 5. As with the single-sided model, the two-sided model (or models) may be trained on any suitable network node and then transmitted to the UE 3 and the base station 5.

[0181] AI / ML model acquisition Next, a method for AI / ML model deployment will be described. In this example, an AI / ML model is sent to UE 3 for use by UE 3. The AI / ML model may be a two-sided model (if a corresponding AI / ML model, or the same AI / ML model, is used by base station 5), or it may be an AI / ML model used only by UE 3.

[0182] In this example, when UE 3 is in an RRC idle state or an RRC inactive state, broadcast or multicast transmission is used to transmit the AI / ML model to UE 3. When UE 3 is in an RRC connected state, multicast transmission and / or an RRC message (e.g., a dedicated RRC message or another new RRC message different from the old RRC message) is used to transmit the AI / ML model to UE 3.

[0183] FIG. 12 illustrates an example of how the base station 5 may broadcast notification of supported AI / ML models.

[0184] In step S1401, the base station 5 broadcasts notification of supported AI / ML models. In this example, the notification of supported AI / ML models is included in system information (SI) broadcast in the base station's cell. The UE 3 in this example is in an RRC idle state or an RRC inactive state (but could alternatively be in an RRC connected state). Thus, advantageously, the UE 3 can receive information indicating which AI / ML models are supported by the base station 5 even when the UE 3 is in an RRC idle state or an RRC inactive state.

[0185] The broadcast SI may include a list of AI / ML model IDs and / or version numbers of supported AI / ML models. The supported AI / ML models may be indicated for each use case. 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 CSI encoding / decoding may also be provided. The notification of supported AI / ML models may be periodically broadcast by the base station 5 or, alternatively, may be broadcast on-demand in response to a request from the UE 3. The broadcast SI may also include notification of how to obtain the AI / ML models (e.g., signaling-based transmission between the UE 3 and the base station 5, or database transmission between the UE 3 and the AI / ML server 151). If the UE 3 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; alternatively, the notification may be transmitted to the UE 3 via multicast transmission.

[0186] In step S1402, the UE 3 determines whether to acquire one of the supported AI / ML models based on the notification of the supported AI / ML models received from the base station 5. In this example, the UE 3 determines to acquire one of the models and, in step S1403, sends a request for the model to the base station 5. Step S1403 may be performed when the UE 3 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 described in more detail below). In step S1404, the base station 5 sends the requested model to the UE 3. As described in more detail below, the UE 3 may be in an RRC connected, RRC inactive, or RRC idle state when it receives the AI / ML model from the base station in step S1404.

[0187] In the example of FIG. 12, the UE 3 sends a request for an AI / ML model to the base station 5 and receives the requested AI / ML model from the base station 5, but this is not necessarily the case. Alternatively, the UE 3 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) after receiving notification of supported AI / ML models. For example, FIG. 13 shows a variation of FIG. 12 in which the UE 3 requests an AI / ML model stored in the AI / ML server 151. FIG. 13 includes new steps S1403b and S1403c. In step S1403b, the base station 5 sends a request for the AI / ML model requested by the UE 3 to the AI / ML server 151. In step S1403c, the AI / ML server 151 sends the requested model to the base station 5 for forwarding to the UE 3 in step S1404. The transfer of the AI / ML model via the base station 5 in steps S1403c and S1404 may be transparent to the base station 5 (e.g., the AI / ML model may be transmitted using one or more transparent containers). In a further alternative, the UE 3 may obtain the AI / ML model from the AI / ML server 151 via the AMF 10-1, for example, using NAS-based signaling. For example, rather than sending a request for the AI / ML model to the base station 5, the UE 3 may send a request for the AI / ML model to the AMF 10-1. The AMF 10-1 may then request the model from the AI / ML server 151 and forward the AI / ML model from the AI / ML server 151 to the UE 3. In another alternative, the UE 3 may request the AI / ML model from the base station 5, and the base station 5 may then request the AI / ML model from the AI / ML server 151. However, rather than transmitting the AI / ML model to the UE 3 via the base station that received the request, the AI / ML model may be transmitted to the UE 3 via the AMF 10-1 (using corresponding NAS signaling).

[0188] When UE 3 requests an AI / ML model stored in AI / ML server 151, the AI / ML model transfer from AI / ML server 151 to UE 3 may be, for example, a normal data transmission, and thus UE 3's acquisition of the AI / ML model from AI / ML server 151 may be transparent to the wireless network from a signaling perspective. However, when UE 3 establishes an RRC connection with the wireless network for such data transmission, UE 3 may include an RRC establishment cause and / or an AI / ML server address (e.g., in the case of AI / ML model transfer) in an RRC message. Base station 5 may forward the information to the core network. Beneficially, this information helps base station 5 and / or core network nodes establish a subsequent user plane data tunnel for AI / ML model transmission between AI / ML server 151 and UE 3.

[0189] The determination of whether to acquire an AI / ML model in step S1402 may be based on a comparison of the AI / ML model stored in the UE 3 with a supported AI / ML model. For example, the base station 5 may provide notification of the model version of the supported AI / ML model in the information broadcast in step S1401, and the UE 3 may compare the version number of the model stored in the UE 3 with the version number of one of the supported models and determine that a newer version of the model is to be acquired. Alternatively, the UE 3 may determine that the UE 3 does not store an AI / ML model for a particular use case (e.g., for encoding CSI) and therefore determine to acquire a supported AI / ML model for that use case. Additionally or alternatively, the UE 3 may determine to send the request for an AI / ML model based on a timer. By using the timer, the UE 3 can request a newer version of the AI / ML model even if the UE 3 has not received the notification of the supported AI / ML model in step S1401 (e.g., the UE 3 may send a request for the latest version of the AI / ML model stored in the UE 3 to the base station 5 based on the timer, regardless of whether the UE 3 received the transmission of step S1401). In a further alternative, the base station 5 can determine to send an updated version of the model to the UE 3 in step S1404 based on the timer. Thus, the base station 5 can provide the UE 3 with a newer version of the AI / ML model even if the base station 5 has not received a request for the newer version of the AI / ML model from the UE 3. This can be particularly beneficial in the case of a two-sided model, where the version of the model in the UE 3 (e.g., for encoding CSI) may need to match or correspond to the version of the model in the base station 5 (e.g., for decoding CSI). By sending a request for an AI / ML model in step S1403 or by sending a transfer of the model in step S1404 based on a timer, the risk of the UE 3 model being inconsistent with the base station 5 model is reduced.It will be appreciated that even if a timer is used for transmitting S1403, UE3 may nevertheless determine to transmit one or more AI / ML mode requests even if the time has not yet expired (e.g., based on information received in step S1401, as described above).

[0190] The UE 3 may perform a random access procedure (such as the RA procedure described above with reference to FIG. 6) to request an AI / ML model from the base station 5. In this example, the MSG 3 sent from the UE 3 to the base station 5 in the RA procedure includes an RRC establishment cause indicating that the UE 3 is requesting the AI / ML model (e.g., by including an indication of the identity of the requested AI / ML model or an indication that the UE 3 should enter an RRC connected state to download the AI / ML model from the base station 5). The UE 3 may request the AI / ML model using the RA procedure in both the example of FIG. 12 , in which the requested AI / ML model is initially stored in the base station 5, and in the method of FIG. 13 , in which the AI / ML model is initially stored in the AI / ML server 151 (or any other suitable network node).

[0191] While the use of MSG3 and an RRC establishment cause provides an efficient mechanism for the UE 3 to indicate that it is requesting an AI / ML model, the notification may alternatively be provided in any other suitable transmission from the UE 3 to the base station 5. For example, the UE 3 may use a new RRC message (e.g., a dedicated RRC message) to indicate that it is requesting an AI / ML model. Any other suitable method of obtaining the AI / ML model may alternatively be used, and the UE 3 does not necessarily have to use an RA procedure to obtain the model.

[0192] In a further alternative, in response to the determination of step S1402, rather than UE 3 requesting the AI / ML model (entering the RRC connected state to receive the model), UE 3 may wait until UE 3 next enters the RRC connected state before obtaining the AI / ML model from base station 5. In another alternative, rather than entering the RRC connected state to receive the AI / ML model, UE 3 may receive the AI / ML model when UE 3 is in the RRC idle state or the RRC inactive state. In this case, base station 5 sends notification of the communication resources (e.g., time and frequency resources) to be used by UE 3 to receive the AI / ML model while UE 3 is in the RRC idle state or the RRC inactive state.

[0193] The AI / ML model may be transmitted from the base station 5 to the UE 3 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., a transmission priority) may be assigned to the DRB or logical channel carrying the AI / ML model. As described above, a logical channel may be assigned (e.g., by the base station 5) an index indicating the priority of the transmission of the logical channel 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 requested (e.g., the AI / ML use case). 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 a higher PBR) than if the AI / ML model is used in a beam prediction procedure.

[0194] From the air interface perspective, if the AIML model transfer undergoes user plane transmission as described above, it may differ from traditional user plane (UP) transmission. Traditional UP transmission requires two or more parts (air interface and backhaul-based fixed network) based transmission, e.g., a DRB over the air interface, plus a data tunnel established between the base station 5 and a UPF in the core network that bridges the data toward a data server. In this traditional method of UP transmission, the base station 5 is not a data producer, but instead a data "consumer," because the base station simply converts one or more QoS flows to a DRB at the SDAP layer to support data transmission for a specific QoS service on a data radio bearer over the air interface. However, this traditional UP transmission can be advantageously modified, and the base station 5 can become a data producer if it itself holds an AIML model ready for transfer to the UE. When the base station 5 determines to transfer the AI / ML model to the UE 3 via the UP-based channel, the 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 data of the AI / ML model is not carried by the SDAP layer, unlike the conventional method.

[0195] Step S1404 of Figures 12 and 13 may include transmitting the AI / ML model to UE 3 using a dedicated radio bearer (e.g., a bearer other than the old SRB / DRB). A new logical channel (e.g., a dedicated logical channel) may be used to transmit the AI / ML model. The logical channel may be assigned a priority and / or PBR as described above; alternatively, the logical channel may simply not be multiplexed with other logical channels and instead may be transmitted separately.

[0196] 14 shows an example of how a requested AI / ML model may be obtained by UE3 when the requested AI / ML model is initially stored in CU60 of a distributed base station. Steps S601 to S603 are the same as steps S1401 to S1403 described above, and therefore will not be repeated here. In step S604, DU50 sends a request for the AI / ML model requested by UE3 to CU60, and in step S605, CU60 sends the AI / ML model to DU50. Step S606, in which DU50 sends the requested AI / ML model to UE3, is the same as step S1404 in FIGS. 14 and 15.

[0197] In step S605, the AI / ML models can be transmitted from the CU 60 to the DU 50 using a new (e.g., dedicated) F1 application protocol (AP) message or procedure. Furthermore, the CU 60 can also transmit to the DU 50 a notification of the supported AI / ML models broadcast by the DU 50 in step S601. The notification of the supported AI / ML models (e.g., model IDs) can be transmitted from the CU 60 to the DU 50 using an F1-AP message (e.g., a dedicated F1-AP message). Thus, the DU 50 can determine the notification of the supported AI / ML models broadcast in step S601.

[0198] As described above, notification of supported AI / ML models may be transmitted in step S1401 (or step S601) using system information broadcast in the cell of the base station 5. A new SIB can be used to transmit notification of supported AI / ML models. This SIB may be referred to as the "AI / ML SIB." SIB1 can be used to provide notification that an AI / ML SIB is available for broadcast within the cell (the AI / ML SIB may be an on-demand SI transmitted in response to a request from the UE 3, not shown in FIG. 12, but transmitted by the UE 3 before step S1401). The MIB and SIB1 can provide the UE 3 with notification of scheduling information for receiving and decoding the dedicated AI / ML SIB. The AI / ML SIB may include model IDs of supported (or "available") AI / ML models. As described above, the AI / ML SIB can indicate the supported AI / ML models for each use case.

[0199] The AI / ML SIB may be broadcast periodically (e.g., according to a predetermined periodic pattern), or alternatively, may be provided "on-demand," for example, in response to a request from UE 3. SIB1 may be used to indicate to UE 3 whether the AI / ML SIB is transmitted periodically or whether it is available on-demand.

[0200] If the AI / ML SIB is available on demand, the base station 5 provides notification of the availability of the AI / ML SIB or information indicating supported AI / ML model IDs for a particular feature (e.g., beam management) in the system information block SIB1. The UE 3 may be configured to request the AI / ML SIB using message 1 (MSG1), which may be referred to as an MSG1-based on-demand SI request for the AI / ML SIB, or message 3 (MSG3), which may be referred to as an MSG3-based on-demand SI request for the AI / ML SIB. The UE 3 may also use another type of uplink message to indicate that the UE 3 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 the UE 3's request for information (e.g., AI / ML IDs), the network broadcasts the supported AI / ML information (e.g., AI / ML model IDs) for the one or more features requested by the UE 3 (e.g., using the system information block AI / ML SIB). The UE 3 can then obtain the AI / ML information by receiving and decoding the broadcasted messages (eg, AI / ML SIBs).

[0201] In the example described above with reference to Figures 12 to 14, UE3 may request a single AI / ML model, or alternatively, may request multiple AI / ML models in step S1403 (or step S603).

[0202] FIG. 15 illustrates a modification of the method of FIG. 12 in which the network is configured to notify one or more UEs 3 of updates to the AI / ML model using paging transmissions.

[0203] In step S701, the base station 5 obtains an updated AI / ML model. The updated AI / ML model may be generated in the base station 5, or the updated AI / ML model may be received from another node in the network (e.g., from the AI / ML server 151 or from a core network node / function). In step S702, the base station transmits a paging transmission including notification that the AI / ML model has been updated. The paging transmission may be a group paging transmission (such as a paging transmission intended for reception by a specific group of UEs 3).

[0204] The paging transmission of step S702 may include notification of where UE 3 should obtain the updated AI / ML model. For example, if the updated AI / ML model is stored in AI / ML server 151, the paging transmission may provide notification that UE 3 will obtain the updated AI / ML model directly from AI / ML server 151 (or from any other suitable network node). The paging transmission may also include notification of the UE 3 that will obtain the updated AI / ML model (e.g., notification of the identity of the UE 3 that will obtain the updated AI / ML model).

[0205] The paging transmission may include a notification that the page is for notification of an updated AI / ML model. For example, the paging transmission may include a cause value indicating that the page is for notification of an updated AI / ML model. The paging transmission may include a notification of the identity of the updated AI / ML model (e.g., a model ID number) and / or a version number of the updated AI / ML model.

[0206] In step S703, the UE 3 determines to acquire an updated AI / ML model based on the information received in step S702. For example, the UE 3 may determine to acquire an updated AI / ML model based on the difference between the version number of the model stored in the UE 3 and the version number of the updated AI / ML model. Alternatively, the UE 3 may determine to acquire an updated AI / ML model based on an explicit notification in the paging transmission of step S702 that the UE 3 should acquire an updated AI / ML model. Steps S704 and S705 are the same as steps S1403 and S1404 described above with reference to FIG. 14, and therefore will not be described again.

[0207] In this example, the paging transmission is used to notify one or more UEs 3 that the AI / ML model has been updated, but alternatively (or additionally) the paging transmission can be used to request the identity of an AI / ML model stored in the UE 3, in which case the UE 3 transmits notification of the AI / ML model stored in the UE 3 to the base station 5 after receiving the request. Alternatively or additionally, the paging transmission can be used to request AI / ML model history information from the UE 3 (e.g., model execution history) or other information regarding the status of the AI / ML model, in which case the UE 3 transmits the AI / ML model history information to the base station 5 after receiving the request.

[0208] Area-based AI / ML models Next, methods related to area-based AI / ML models are described. An AI / ML model may be for use in a particular area or location. An AI / ML model may be for use in a particular cell or group of cells that may be operated by one or more base stations 5. For example, an AI / ML model may be for use in a group of cells for beam management.

[0209] The area 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 will be understood that any other suitable area for using an AI / ML model may be defined. An area in which an AI / ML model is used for a particular function (e.g., beam management, CSI encoding / decoding, or mobility) may be referred to as an AI / ML model functional area.

[0210] A cell provided by a base station 5 may be part of multiple AI / ML model functional areas. FIG. 16 shows an example in which 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 182. In this example, a first AI / ML model is for use in the first cell 180 and the second cell 181 with respect to a first function (e.g., beam management). Thus, the AI / ML model functional area of ​​the first AI / ML model includes the first cell 180 and the second cell 181. A second AI / ML model is for use in the second cell 181 and the third cell 182 with respect to a second function (e.g., CSI encoding / decoding). Thus, the AI / ML model functional area of ​​the second AI / ML model includes the second cell 181 and the third cell 182.

[0211] In this example, the second cell 181 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. In this example, one AI / ML model is used for each area function. Alternatively, more than one AI / ML model may be available for use for a particular area function (e.g., more than one AI / ML model may be available for UE mobility inference in a particular cell).

[0212] In this example, the base station 5 is configured to send a broadcast transmission on a first cell 180 indicating that the first cell 180 belongs to an AI / ML model functional area of ​​a first AI / ML model, and to send a broadcast transmission on a 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 ​​a second AI / ML model. The indication of which AI / ML model functional area a cell belongs to may be referred to as AI / ML model area information. Thus, a UE 3 in the cell of the base station 5 can determine which AI / ML model to use for a particular function in that cell. The base station 5 may be configured to indicate, in the broadcast transmission, the model functional area to which the cell belongs, for each AI / ML model or for each function. For example, the base station 5 may support two AI / ML features / functions, where AI / ML model X is used for the first function and AI / ML model Y is used for the second function. From a network deployment perspective, an AI / ML model X of a first feature may belong to area N (which may be, for example, a relatively small area), and an AI / ML model Y of a second feature may belong to area M (which may be, for example, a relatively large area). The broadcast information may indicate that a cell supports AI / ML models X and Y, may indicate that a cell supports a first feature with AI / ML model X and a second feature with AI / ML model Y, or may alternatively indicate that a cell is part of corresponding areas N and M (for different models or features).

[0213] 17 is a diagram showing a method in which AI / ML model area information is received by UE 3. In step S1901, base station 5 transmits (broadcasts) AI / ML model area information in the cell of base station 5, and the information is received by UE 3 in the cell.

[0214] In step S1902, the UE 3 determines to use a particular AI / ML model based on the AI / ML model area information. For example, if the UE 3 is located in the second cell 181 of FIG. 16 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, the UE 3 determines to use the first AI / ML model for the first function of the second cell 181. If the UE 3 does not support the AI / ML model indicated in the AI / ML model area information, the UE 3 may simply ignore the AI / ML model area information. After the UE 3 determines to use a particular AI / ML model, the UE 3 may acquire the AI / ML model (if the AI / ML model is not already stored in the UE 3) according to any of the methods described herein (e.g., any of the methods shown in FIGS. 12 through 15). For example, the UE 3 may use a random access procedure including MSG3 as part of the method for acquiring the AI / ML model, as described above. As mentioned above, the UE 3 may obtain the AI / ML model directly from the base station 5 or from another node in the network (e.g., from the AI / ML server 151 (via the AMF 10-1), from an operations, administration, and maintenance server (OAM), or from any other suitable node / function in the network).

[0215] Alternatively, the AI / ML model area information may simply include a notification that a particular function is supported in the area, rather than the AI / ML model area information including a notification of which AI / ML models are supported for a particular function. In this case, after receiving the AI / ML model area information, the UE 3 may determine to acquire system information broadcasted in the cell to determine which AI / ML model to use. For example, as described above, the UE 3 may request an on-demand SIB including a notification of the AI / ML models supported for a particular function in the cell. The UE 3 may determine to acquire an AI / ML model after moving to a new cell (e.g., following a cell reselection procedure) and receiving the broadcast transmission of step S1901, or if the AI / ML model to use for a particular function in the cell has changed (which the UE 3 may also identify based on the broadcast transmission of step S1901). The UE 3 may be configured to periodically check the transmission of the AI / ML model area information by the base station 5 (e.g., by receiving and decoding the corresponding SI) based on a timer. Similarly, the base station 5 may be configured to periodically broadcast the AI / ML model area information in one or more cells based on a timer.

[0216] If the AI / ML models supported for use with functions in a particular area are updated, UE3 may obtain the updated models using any of the methods described above (e.g., the method described with reference to FIG. 15).

[0217] The UE 3 may store multiple AI / ML models that can be used for a specific function, and the UE 3 may select one of the multiple AI / ML models based on the AI / ML model area information received in step S1901. For example, the UE 3 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 a notification in the AI / ML model area information that the cell belongs to the first area. If the cell belongs to both the first area and the second area, the UE 3 may provide a notification to the network (e.g., to the base station 5) of whether to use the first AI / ML model or the second AI / ML model (e.g., which model is preferred by the UE 3). Thus, advantageously, in the case of a dual-sided model, a mismatch between the model used by the base station 5 and the model used by the UE 3 can be avoided when multiple models are supported for the same function in a specific area. The UE 3 may provide an indication of which AI / ML model to use (or which AI / ML model is preferred to use) using the first RRC message sent after the UE 3 to the base station 5. The UE 3 may include the indication in MSG 3 described above with reference to FIG.

[0218] In the example of FIG. 16 , mobility of UE 3 may occur from the second cell 181 to the third cell 182. Meanwhile, in the second cell 181, UE 3 uses a first AI / ML model for a first function. However, in this example, the third cell 182 does not support the first function. Therefore, UE 3 may determine not to use (or disable) the first AI / ML model for the first function after UE 3 moves to the second cell 181. For example, UE 3 may determine 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 an AI / ML model supported in the third cell 182 (or the AI / ML model functional area to which the third cell 182 belongs). Also, UE 3 may determine not to use (or disable) the first AI / ML model in the third cell 182 even if UE 3 does not receive a broadcast transmission from the second base station 5-2. For example, the UE 3 may determine not to use (or disable) the first AI / ML model for the third cell 182 as a default option, and may only determine to use the first AI / ML model for the third cell if the UE 3 receives notification that the first AI / ML model is available for use in the third cell 182. Thus, advantageously, the use of an unsupported AI / ML model or AI / ML model function can be avoided even if the second base station 5-2 is an older base station that may not support transmission of AI / ML-related information, such as the transmission of step S1901 of FIG.

[0219] Model Updates and RRC State Transitions When UE 3 transitions from an RRC idle state or an RRC inactive state to an RRC connected state, UE 3 can use layer 1 (L1), layer 2 (L2), or layer 3 (L3) signaling to indicate to the network which AI / ML models are stored at UE 3 (e.g., by sending associated AI / ML model IDs). UE 3 can also use L1 / L2 / L3 signaling to provide notification to the network of the versions of AI / ML models stored at UE 3. L1 / L2 / L3 signaling can also be used to provide notification of historical information related to AI / ML models (e.g., model execution history).

[0220] Based on the L1 / L2 / L3 signaling, the network (e.g., base station 5) can determine a particular AI / ML model to be used for a particular function. For example, based on the L1 / L2 / L3 signaling, base station 5 can determine that UE 3 has stored an AI / ML model that is also supported by base station 5, and therefore can determine to use the AI / ML model for a particular function (e.g., beam management, or CSI encoding / decoding). If the AI / ML model is not already stored in UE 3, base station 5 can determine to send the AI / ML model to UE 3, or can determine to send a different version of the AI / ML model stored in UE 3 to UE 3. Base station 5 can send the AI / ML model to UE 3 following UE 3's transition to an RRC connected state (e.g., immediately after UE 3 transitions to an RRC connected state).

[0221] To avoid mismatches between the AI / ML model used by the UE 3 and the AI / ML model used by the base station 5 (in the case of a two-sided AI / ML model), the base station may be configured not to use the AI / ML model until the AI / ML model is transmitted to the UE 3 or until the base station 5 receives an acknowledgment from the UE 3 that it has acquired the AI / ML model. For example, the base station 5 may use a non-AI / ML algorithm for CSI compression / decompression. The base station 5 may use a DCI or medium access control (MAC) control element (CE) to control activation of the use of the AI / ML model in the UE 3 (e.g., for a particular function).

[0222] The base station 5 may also receive in L1 / L2 / L3 signaling information indicative of the performance of the AI / ML model used in the UE 3. The model performance information may for example be the model performance feedback of Figure 8 or may be the information used in the performance monitoring step of Figure 9.

[0223] When the UE 3 transitions from the RRC connected state to the RRC idle state or the RRC inactive state, the UE 3 may be configured to continue to store one or more AI / ML models stored in the UE 3. The UE 3 may be configured to continue to store the AI / ML models for a predetermined period of time, for example, based on a timer. However, it will be appreciated that if the UE 3 receives additional AI / ML models and does not have sufficient memory to store both models, the UE 3 may be configured to delete or overwrite the AI / ML models stored in the memory of the UE 3. It will be appreciated that if the UE 3 is configured to continue to store one or more AI / ML models after the UE 3 transitions from the RRC connected state to the RRC idle state or the RRC inactive state, the AI / ML models are not part of the RRC connected state UE context because the RRC connected state UE context is deleted after the UE transitions from the RRC connected state to the RRC idle state or the RRC inactive state.

[0224] RRC Procedure RRC procedures may be used for AI / ML-related queries sent between the network and the UE 3 when the UE 3 is in an RRC connected state. For example, the network may request (e.g., via the base station 5) information indicating the identity of one or more AI / ML models stored in the UE 3 using an RRC message (e.g., a dedicated RRC message or other non-stale RRC message). The network may request information indicating the identity of one or more AI / ML models stored in the UE 3 for a particular function or feature. The UE 3 may send a corresponding RRC message including the requested information to the base station 5. For example, the UE 3 may send an RRC message including an indication of the AI / ML model IDs of the AI / ML models stored in the UE 3 to the base station 5.

[0225] Similarly, UE 3 may request AI / ML-related information from the network (e.g., via base station 5) using an RRC message (e.g., a dedicated RRC message or other non-stale RRC message). For example, UE 3 may request the identities of AI / ML models supported by base station 5 for a particular function, or may request version numbers of AI / ML models available at base station 5 (e.g., UE 3 may request the current version number of an AI / ML model to obtain the latest version of the model). Base station 5 may then send a corresponding RRC message to UE 3 containing the requested information (e.g., including an indication of the AI / ML model IDs of the AI / ML models stored at base station 5).

[0226] AI / ML Model Transfer Next, we will describe an improved method for providing AI / ML models to UE3. FIG. 18 is a diagram illustrating how the AI / ML model is transmitted by the base station 5 to the UE 3. In step S1801, the base station 5 transmits system information indicating AI / ML features (sometimes referred to as "use cases") to be implemented (e.g., activated) in the UE 3. The AI / ML features may utilize AI / ML models for one or more of energy conservation, traffic steering, anomaly detection, quality of experience (QoE) optimization, mobility robustness optimization (MRO), RAN slice service level agreement (SLA) guarantees, massive 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. In this example, the system information is broadcast by the base station 5 within its cell. However, this need not necessarily be the case. The base station may transmit notification of the AI / ML features implemented by the UE 3 in any other suitable transmission, such as a multicast transmission. The system information S1801 may be on-demand system information or may be dedicated system information to indicate the AI / ML features implemented in the UE 3.

[0227] The system information transmitted in step S1801 may include information indicating how the UE 3 can obtain the corresponding AI / ML model information. For example, the system information may include a notification of a request message transmitted by the UE 3 to obtain the AI / ML model information and corresponding scheduling information. The system information transmitted in step S1801 may also include a notification of information that the UE 3 should include in the request to obtain the AI / ML model information. For example, the system information may include a notification that the UE 3 will transmit its vendor information to obtain the AI / ML model information, or the UE 3's AI / ML capability information (e.g., AI / ML models or features supported by the UE 3) in the request. Alternatively, the base station 5 can simply transmit the AI / ML model information in the system information transmitted in step S1801 (which may also include a notification of communication resources used by the UE 3 to obtain an AI / ML model for implementing the AI / ML feature).

[0228] In optional step S1802, the UE 3 sends a request for AI / ML model information to the base station 5 if the UE 3 has not yet received the AI / ML model information from the base station in step S1801. The request for AI / ML model information may include notification of model parameters supported by the UE 3 (e.g., by indicating the capabilities of the UE 3), or other information (e.g., characteristics of the UE 3, such as the type of UE 3, the model of the UE 3 (e.g., model number), or the vendor of the UE 3) used by the base station 5 to determine the AI / ML model (and / or corresponding AI / ML model parameters) to be sent to the UE 3.

[0229] In step S1803, the base station 5 transmits AI / ML model information to the UE 3. The AI / ML information includes notification of the AI / ML model acquired by the UE 3 (alternatively, this information may be provided in step S1801, in which case step S1803 does not necessarily need to be performed). The transmission in step S1803 may include notification of the identity of the AI / ML model acquired by the UE (e.g., a model ID number), configuration parameters of the AI / ML model, and / or notification of the size of the AI / ML model (e.g., the memory required by the UE 3 to store the model). As will be described below with reference to FIG. 20, if the UE 3 acquires the AI / ML model from the server 151 or other network node, the AI / ML model information transmitted in step S1803 also includes information for the UE 3 to use to acquire the model from the server 151 or other network node (e.g., an internet protocol, IP address, or server name).

[0230] In step S1804, the UE 3 determines to acquire an AI / ML model based on the information received in step S1803. For example, the UE 3 may determine whether the UE 3 already stores the AI / ML model indicated in the transmission of step S1803, and may determine to acquire the model if the UE 3 does not yet store the model. Alternatively, for example, the UE 3 may determine to acquire the model if the version of the model indicated in step S1803 is newer than the version of the model stored in the UE 3.

[0231] In step S1805, the UE 3 sends a request for an AI / ML model. The AI / ML request includes an indication of the requested model (e.g., the model ID number of the model, or the version number of the model).

[0232] In step S1806, the base station 5 transmits AI / ML model transmission information to the UE 3. The AI / ML model transmission information (which may also be referred to as AI / ML model transfer configuration information) includes notification of communication resources to be used when transmitting the AI / ML model to the UE 3 (e.g., notification of time and / or frequency resources, e.g., the configuration of one or more resource blocks). The AI / ML model transmission information may include notification of a radio bearer configuration for the transmission of the AI / ML model from the base station 5 to the UE 3. The radio bearer for the transmission of the AI / ML model may be a data radio bearer, a DRB, without UPF involvement, or may be a dedicated SRB for the transfer of the model. The AI / ML model transmission information may include scheduling information (in the time and / or frequency domain) for transmitting the model to the UE 3. The scheduling information may include notification of whether the model transfer is to be performed using the next available transmission or after a configured delay (e.g., based on a timer). Alternatively, the AI / ML model transmission may be triggered by a subsequent transmission from the base station (as described with reference to optional step S1807). The base station 5 may delay transmitting the AI / ML model to the UE 3, for example, because the AI / ML model is also transmitted to one or more other UEs 3 in a multicast transmission and the other UEs 3 are not yet configured to receive the multicast transmission.

[0233] In optional step S1807, the base station 5 transmits a notification that the model transfer is initiated. The transmission of step S1807 may be a multicast transmission, for example, if the AI / ML model is transmitted from the base station 5 to multiple UEs 3, or a paging transmission (e.g., a group paging transmission). The transmission of step S1807 may include notification of time resources used by the base station 5 to transmit the AI / ML model to the UEs 3. Alternatively, step S1807 may be omitted, and the method may proceed to step S1808 after step S1806.

[0234] In step S1808, the AI / ML model is transmitted from the base station 5 to the UE 3 using the communication resources indicated in step S1806. It will be appreciated that if the base station 5 is a distributed base station, the model may be stored in the CU-UP or CU-CP of the base station 5 and transmitted to the UE 3 via the DU 50 of the base station (and may be transmitted from the CU-UP or CU-CP to the DU 50 via the F1-U or F1-C interface, respectively).

[0235] In optional step S1809, the UE 3 sends an acknowledgement (ACK) to the base station 5 indicating that the AI / ML model has been received by the UE 3. If the model has not been successfully received by the UE 3, the UE 3 sends a negative acknowledgement (NACK) to the base station 5, and the base station 5 may determine to resend the AI / ML model to the UE 3. Alternatively, the UE 3 may simply not send an ACK to the base station 5 if the UE 3 has not successfully received the AI / ML model.

[0236] In step S1810, the UE 3 may activate the received AI / ML model and start using the AI / ML model for the corresponding use case or feature. The UE 3 may determine to activate the model after completing reception of the model from the base station 5. Alternatively, the base station 5 may provide a notification that the model should be activated after the model is received at the UE 3 (e.g., after the UE 3 sends an ACK in step S1809). Sending a notification that the model should be activated by the base station 5 helps ensure synchronization between the model used by the UE 3 and the model used by the base station 5 when a two-sided model is used. Alternatively, the UE 3 may provide a notification to the base station 5 that the UE 3 has activated the AI / ML model.

[0237] 19 is a diagram illustrating a variation of the example of FIG. 18, in which the AI / ML model is transmitted from the base station 5 to the UE 3 without providing a separate notification of communication resources for transmitting the AI / ML model. In this example, the base station 5 responds to a request for the AI / ML model by transmitting the AI / ML model. Steps S191 to S195 are the same as steps S1801 to S1805, and therefore will not be described again here.

[0238] In step S196, the base station 5 transmits the AI / ML model to the UE 3. The model may be transmitted to the UE 3, for example, using an RRC message or using any other suitable transmission. Steps S197 and S198 are the same as steps S1809 and S1810, respectively, and therefore will not be described again here.

[0239] FIG. 20 illustrates a variation of the example of FIG. 18 in which the AI / ML model is sent from the AI / ML server 151 to the UE 3. Steps S201 and S202 correspond to steps S1801 and S1802 in FIG. 18, and therefore will not be described again here.

[0240] Step S203 corresponds to step S1803 in Figure 18, except that in step S203, the AI / ML model information includes information for use by UE 3 to obtain the model from AI / ML server 151 (e.g., internet protocol, IP, address, server name, or configuration information of one or more bearers). Alternatively, this information may be provided to UE 3 from the base station in step S201, in which case step S203 does not necessarily need to be performed.

[0241] Step S204 is the same as step S1804 in FIG. 18, so a description thereof will be omitted here. In step S205, similar to step S1805 in FIG. 18, the UE 3 transmits a request for an AI / ML model to the AI / ML server 151.

[0242] In step S206, the AI / ML model is transmitted from the AI / ML server 151 to the UE 3. The AI / ML model can be transmitted directly from the AI / ML server 151 to the UE 3, or can be transmitted from the AI / ML server 151 to the UE 3 via another network node (e.g., via the base station 5 as shown in FIG. 13 or via the AMF 10-1).

[0243] In optional step S207, the UE 3 sends an acknowledgement (ACK) to the AI / ML server 151 indicating that the AI / ML model has been received by the UE 3. Alternatively, the ACK can be sent to the base station, as illustrated in step S1809 of FIG. 18. If the model is not successfully received by the UE 3, the UE 3 sends a negative acknowledgement (NACK) to the AI / ML server 151 (or to the base station 5), and the AI / ML model is retransmitted to the UE 3. Alternatively, rather than sending a NACK, the UE 3 may simply not send an ACK if it does not successfully receive the AI / ML model.

[0244] Step S208 is the same as step S1810 in FIG. 18, so a description thereof will be omitted here. 20, the model is stored in the AI / ML server 151 and transmitted from the AI / ML server 151 to the UE 3, but the network node that stores the model and transmits it to the UE 3 does not necessarily have to be the AI / ML server 151. Alternatively, for example, the network node may be a service management and orchestration (SMO) entity or a RAN intelligent controller (RIC) (e.g., a non-real-time RIC).

[0245] 20, the AI / ML model may be transmitted from the AI / ML server 151 to the UE 3 using local IP access (LIPA) or selected IP traffic offload (SIPTO). Beneficially, these methods allow data to be transferred over a local network (e.g., the Internet, via a local gateway, etc.) rather than over the core network of the communications system 1. We now describe a variation of the example shown in Figures 18 to 20 in which the UE 3 sends notifications of one or more supported AI / ML models or features to the network.

[0246] 21 is a diagram illustrating an example in which a UE 3 transmits UE AI / ML capability information to a base station 5 and receives an AI / ML model from the base station. In this example, the transmission of the AI / ML model to the UE 3 is initiated by the network.

[0247] In step S210, the UE 3 transmits UE AI / ML capability information to the base station 5. The UE AI / ML capability information indicates the AI / ML capabilities of the UE 3. For example, the UE AI / ML capability information may include an indication of one or more AI / ML features (or use cases) supported by the UE 3, or an indication of one or more AI / ML models (e.g., AI / ML model IDs or version numbers) supported or stored by the UE 3. This method can be used, for example, when the UE 3 stores an AI / ML model, but the base station 5 does not broadcast an indication that the model is supported for use in the base station 5's cell.

[0248] In optional step S211, the base station sends a request for further UE AI / ML capability information to the UE 3. For example, if not received in step S210, the base station 5 may send a request for notification of AI / ML models supported or stored by the UE 3. In optional step S12, the UE 3 sends the requested information to the base station 5.

[0249] Alternatively, rather than the base station 5 sending a request for further UE AI / ML capability information in step S211, the base station 5 may send a request to the UE 3 to activate a particular AI / ML model. If the model is available (e.g., stored) in the UE 3, the UE 3 activates the AI / ML model and sends a notification to the base station that the AI / ML model has been activated. If the model is not available in the UE 3 or is not supported by the UE 3, the UE 3 may send a notification that the AI / ML model is not stored and / or supported in the UE 3 (e.g., an error message including a notification of the reason why the UE 3 cannot activate the AI / ML model requested by the base station 5, such as a cause value). If the AI / ML model requested to be activated by the base station 5 is not available in the UE 3 or is not supported by the UE 3, the UE 3 may send a notification to the base station of one or more AI / ML models available for use by the UE 3.

[0250] After receiving the UE AI / ML capability information in step S210 (and optionally in step S212), the base station 5 determines whether to transmit an AI / ML model to the UE 3. For example, if an AI / ML model is supported by the base station 5 and the UE 3 for a particular feature, but the UE 3 does not currently store the AI / ML model, the base station 5 may determine to transmit the AI / ML model to the UE 3. The method then proceeds to step S213, where the base station transmits AI / ML model transmission information to the UE 3. The AI / ML model transmission information (which may also be referred to as AI / ML model transfer configuration information) includes notification of communication resources to be used in transmitting the AI / ML model to the UE 3 (e.g., notification of time and / or frequency resources, e.g., configuration of one or more resource blocks). The AI / ML model transmission information may include notification of a radio bearer configuration for transmission of the AI / ML model from the base station 5 to the UE 3. The radio bearer for transmission of the AI / ML model may be a data radio bearer, DRB, without UPF involvement, or may be a dedicated SRB for model transfer. The AI / ML model transmission information may include scheduling information (in the time and / or frequency domain) for transmitting the model to the UE3.

[0251] In optional step S214, the UE 3 transmits an AI / ML model transmission information acknowledgement message to the base station 5 indicating that the UE 3 has received the information transmitted in step S213. For example, if the transmission in step S213 includes notification of a data radio bearer (DRB) for use by the UE 3 to receive the AI / ML model, the UE 3 may transmit in step S214 a notification that the UE 3 has received the configuration of the (DRB).

[0252] In step S215, the UE 3 receives the AI / ML model from the base station 5 using the communication resources indicated in step S213.

[0253] In step S216, the UE 3 sends an acknowledgement (ACK) to the base station 5 indicating that the AI / ML model has been received by the UE 3. If the model has not been successfully received by the UE 3, the UE 3 sends a negative acknowledgement (NACK) to the base station 5, and the base station 5 can determine to resend the AI / ML model to the UE 3. Alternatively, the UE 3 may simply not send an ACK to the base station 5 if the UE 3 has not successfully received the AI / ML model.

[0254] In step S217, the base station 5 transmits an AI / ML model activation request to the UE 3. The AI / ML model activation request includes a notification that the UE 3 should activate the AI / ML model transmitted in step S215. Alternatively, the AI / ML model activation request may include a notification that the UE 3 should activate an AI / ML model other than the AI / ML model transmitted in step S215 (e.g., the model notified by the UE 3 is stored in the UE 3 in step S210 or step S212). In this example, the AI / ML model activation request transmitted in step S217 is transmitted separately from the transmission of the AI / ML model itself in step S215, but this is not necessarily the case. Alternatively, the AI / ML model activation request may be transmitted together with the AI / ML model in step S215.

[0255] In step S218, the UE 3 sends a corresponding AI / ML model activation response to the base station 5. If the AI / ML model is available for activation at the UE 3 (e.g., because the AI / ML model was successfully received in step S215), the UE 3 sends a notification to the base station 5 that the UE 3 should activate the AI / ML model. In step S219, the UE activates the AI / ML model. Alternatively, step S219 may be performed before step S218, whereby the UE 3 activates the AI / ML model and then sends a notification to the base station 5 that the AI / ML model has been activated.

[0256] FIG. 22 illustrates an example in which a UE 3 transmits UE AI / ML capability information to a base station 5, and an AI / ML model is transmitted from the base station to the UE without providing a separate notification of communication resources for AI / ML model transmission.

[0257] Steps S220 to S222 are the same as steps S210 to S212 in FIG. 21, and therefore will not be described again here. In step S223, the base station 5 transmits the AI / ML model to the UE 3. The model may be transmitted to the UE 3, for example, using an RRC message or using any other suitable transmission. Steps S224 to S227 are the same as steps S216 to S219 in FIG. 21, and therefore will not be described again here.

[0258] FIG. 23 is a diagram showing an example in which the UE 3 transmits UE AI / ML capability information to the base station 5, and an AI / ML model is transmitted from the AI / ML server to the UE. Steps S230 to S232 are the same as steps S210 to S212 in FIG. 21, and therefore will not be described again here.

[0259] In step S233, the base station transmits AI / ML model transmission information to the UE 3. The AI / ML model transmission information includes notification of information for use by the UE 3 to obtain a model from the AI / ML server 151 (e.g., internet protocol, IP, address, server name, or configuration information of one or more bearers). The AI / ML model transmission information may also include notification of the AI / ML model that the UE 3 obtains from the AI / ML server 151 (e.g., AI / ML model ID number, or version number). In step S234, the UE3 sends a corresponding acknowledgement that the UE3 has received the AI / ML model transmission information.

[0260] In step S235, the UE3 sends a request for an AI / ML model to the AI / ML server 151 based on the information received in step S233. In step S236, the AI / ML model is transmitted from the AI / ML server 151 to the UE 3. The AI / ML model can be transmitted directly from the AI / ML server 151 to the UE 3, or can be transmitted from the AI / ML server 151 to the UE 3 via another network node (e.g., via the base station 5 as shown in FIG. 13 or via the AMF 10-1).

[0261] In optional step S237, the UE 3 sends an acknowledgement (ACK) to the AI / ML server 151 indicating that the AI / ML model has been received by the UE 3. Alternatively, the ACK can be sent to the base station, as illustrated in step S1809 of FIG. 18. If the model is not successfully received by the UE 3, the UE 3 sends a negative acknowledgement (NACK) to the AI / ML server 151 (or to the base station 5), and the AI / ML model is resent from the AI / ML server to the UE 3. Alternatively, rather than sending a NACK, the UE 3 may simply not send an ACK if it does not successfully receive the AI / ML model. Steps S238 to S240 are the same as steps S217 to S219 in FIG. 21, and therefore will not be described again here.

[0262] Signaling and Model Transmission Next, we will describe the protocols, layers, and messages that may be used for the transmissions illustrated in FIGS.

[0263] RRC Layer Signaling and Transmission Model The RRC layer may be used for any / all of the transmissions shown in Figures 18 to 23. For example, all of the transmissions in steps S1801 to S1803 and S1805 to S1809 may be RRC transmissions / messages, including the transmission of the AI / ML model in step S1808. However, the inventors recognized that if an RRC transmission is used to transmit the AI / ML model in step S1808, it is advantageous to use an RRC message with more segmentation because the amount of data required to transmit the AI / ML model is relatively large. RRC segments are described in more detail in 3GPP TS 38.331. The RRC message may be modified to increase the number of allowed segments, or dedicated procedures may be used to handle segmentation in the case of AI / ML model transfer. A new (e.g., dedicated) radio bearer may also be used for the transfer of the AI / ML model, which advantageously allows improved control by the network of the model's transmission. For example, a dedicated radio bearer allows for improved quality of service (QoS) provisioning and improved retransmission control (e.g., to account for the relatively large data size for the transmission of AI / ML models). The PDCP and RLC functions may also be modified to accommodate the relatively large data size for the transmission of AI / ML models (e.g., by changing the PDCP segmentation method).

[0264] Dedicated AI / ML protocol layer A new (e.g., dedicated) protocol layer may be used for any / all of the transmissions shown in Figures 18 through 23. For example, all of the transmissions in steps S1801 through S1803 and S1805 through S1809 may be transmissions of a layer dedicated to AI / ML related transmissions, including the transmission of the AI / ML model in step S1808.

[0265] In other words, a dedicated AI / ML protocol layer may be used for any / all of the transmissions shown in Figures 18 through 23. Advantageously, the use of a dedicated AI / ML protocol layer allows for more efficient transmission of AI / ML models, for example, by allowing for more message segments or by providing dedicated radio bearer configurations. The RRC layer may be used to set up radio bearers used for signaling and / or model transfer between the UE 3 and the network for dedicated AI / ML protocol layers.

[0266] RRC layer and dedicated AI / ML protocol layer A combination of the RRC layer and a dedicated AI / ML protocol layer can be used for the transmissions shown in Figures 18 to 23. For example, the dedicated AI / ML protocol layer can be used for transmitting the model (e.g., in step S1808), and the RRC layer can be used for all remaining transmissions. Advantageously, this provides a more efficient way to transmit the model itself (e.g., by allowing more message segments or by providing a dedicated radio bearer configuration), while utilizing the RRC layer for other transmissions.

[0267] RRC layer and application layer A combination of the RRC layer and the application layer (layer 7, L7) can be used for the transmissions shown in Figures 18 to 23. For example, the application layer can be used to transmit the model (e.g., in step S1808), and the RRC layer can be used for all remaining transmissions. Advantageously, this provides a more efficient way to transmit the AI / ML model while utilizing the RRC layer for other transmissions. The RRC layer can be used to provide support information for the model transfer (e.g., a transport address to be used by the application layer for model transfer and notification of which AI / ML model should be transmitted).

[0268] Protocol Stack A new (e.g., proprietary) protocol stack may be used for any / all of the transmissions shown in Figures 18 through 23. The protocol stack may support the old PDCP, RLC, MAC, and PHY functions, and may support a proprietary protocol layer for the transmission of AI / ML models. The establishment and release of the protocol stack between the UE 3 and the base station 5 may be achieved using new (eg, dedicated) RRC messages.

[0269] In the protocol stack, a PDCP function (or a PDCP-like function) can use a security key specific to the AI / ML model to encrypt and decrypt the model data. The security key may be a common key and may be applied to each UE 3 that receives the AI / ML model.

[0270] A data radio bearer may be used for transmitting the AI / ML model, the bearer terminating at the UE 3 and the base station 5. Alternatively, the AI / ML mode transmission may be performed between peer protocol entities (e.g., dedicated AI / ML model transmission protocol layers), in which case a data radio bearer does not need to be used for transmitting the AI / ML model.

[0271] In this example, resource allocation may be performed (e.g., by base station 5) for the transfer of the AI / ML model, and therefore, advantageously, data for transmission of the AI / ML model does not need to be multiplexed with other data transmissions to UE 3.

[0272] Data segmentation for transmission of AI / ML models can be performed by a dedicated AI / ML protocol layer or using PDCP functionality (or PDCP-like functionality, since AI / ML model sizes can be larger than the largest available PDCP PDU). A larger PDCP PDU size can be used for transmission of AI / ML models (e.g., 64 MB, or any other suitable size).

[0273] The protocol stack may be released if the connection between the UE 3 and the base station 5 is interrupted and the transmission of the AI / ML model to the UE 3 is not completed. In this case, the UE 3 may maintain (continue to remember) the context associated with the protocol stack, for example, until a timer expires. Thus, advantageously, the UE 3 can use the maintained context for the protocol stack when the UE 3 reconnects to the network (and transmit the maintained context to the network), even if the UE 3 is connected to a different base station 5 than the base station 5 to which the UE 3 was connected when the connection was interrupted (in which case the new base station 5 may continue transmitting the AI / ML model, if available at the new base station 5).

[0274] AI / ML model transmission interruption We now describe an improved method for dealing with scenarios where the transmission of an AI / ML model to a UE3 is interrupted. Since the amount of data transmitted when transmitting the AI / ML model to the UE 3 is relatively large, it takes a relatively long time to transmit the AI / ML model to the UE 3. Therefore, there is a high risk that the transmission will be interrupted. This interruption may occur, for example, due to a radio link failure (RLF) or a handover of the UE 3 to another base station 5.

[0275] If an RLF occurs during transmission of an AI / ML model to UE 3, UE 3 may be configured to discard segments of the AI / ML model (or other units of received data) before the RLF, and transfer of the model will resume after UE 3 reconnects to the network.

[0276] If the UE 3 is communicating with the base station 5 via a cell for transferring the AI / ML model when the RLF occurs and communication with the same base station 5 via the same cell is subsequently restored, the UE 3's model transfer can be resumed. In this case, the UE 3 may transmit model transmission context information to the base station 5 indicating the status of the transmission of the AI / ML model. For example, the model transmission context information may include an indication of the number of segments (or any other suitable data unit) of the AI / ML model received by the UE 3, an indication of the last segment received by the UE 3, or any other suitable information. The base station 5 can advantageously use the transmission context information to configure the transmission of the remaining portions of the AI / ML model to the UE 3. If the UE 3 connects to a different base station 5 or to the same base station 5 via a different cell, the UE can discard (e.g., delete, overwrite, or configure to be overwritten) the segments of the AI / ML model received before the RLF, and the model transfer is restarted (if the AI / ML model is available for transmission).

[0277] If the UE 3 communicates with the base station 5 through a cell for transfer of the AI / ML model when an RLF occurs and then connects to the network through a different cell of the base station 5 or through a different base station 5, transmission of the model can resume if the AI / ML model is also available for transmission by the new base station 5 or is available for transmission in the new cell. If the cell with which the UE 3 is communicating changes following an RLF, the UE 3 can provide a notification to the network (e.g., base station 5) of the cell that was used to transmit the AI / ML model before the RLF. If the new cell is provided by the same base station 5 that provided the original cell, the base station 5 can determine the status of the AI / ML model transmission (e.g., the last segment or other data transmission unit successfully transmitted to the UE 3). If the new cell is provided by a different base station 5 than the base station that provided the original cell, the new base station 5 can request the status of the AI / ML model transmission from the first base station 5 (e.g., via the Xn interface), and the original base station transmits the requested status to the new base station 5. Thus, advantageously, the new base station 5 can determine the status of the AI / ML model transmission and resume transmission of the AI / ML model to the UE 3 following a change in RLF and base station 5.

[0278] The UE 3 may be configured to attempt to resume transfer of the AI / ML model to the UE 3, even if there is a change of cell or base station 5 following an RLF. Following an RLF, the UE 3 may send a notification to the base station 5 (which may be a different base station 5 from the base station 5 from which the UE 3 received the AI / ML model before the RLF) that transmission of the AI / ML model has been suspended and a notification of the status of the AI / ML model transmission (e.g., the last segment or other data transmission unit successfully transmitted to the UE 3). If the base station 5 or cell changes following an RLF and the AI / ML model is not available for transmission (e.g., because the AI / ML model is not stored in the new base station 5), the UE 3 may send a notification that the request to resume AI / ML transmission has been denied.

[0279] The transmission of the AI / ML model to the UE 3 may be via RRC transmission or user plane (UP) transmission. When the AI / ML model is transmitted to the UE 3 using one or more RRC messages, the segments may be RRC segments, each having an associated RRC segment number. The RRC segment number of the last segment received by the UE 3 may be stored by the UE 3 (e.g., in the RRC context information). The RRC segment number of the last segment received by the UE 3 may be stored by the base station 5 (e.g., in the RRC context information). Advantageously, when the UE 3 experiences RLF, the UE 3 and the base station 5 can maintain RRC context information including an indication of the last segment received by the UE 3. Thus, once the RRC connection between the UE 3 and the base station 5 is resumed, transmission of the AI / ML model can be resumed based on the stored RRC context information. Alternatively, the UE 3 may transmit the segment number of the last received RRC segment (and optionally the AI / ML model ID of the model that was being transmitted) as part of the RLF report during the RRC re-establishment procedure. The base station 5 can then advantageously resume transmitting the AI / ML model based on the notified segment number.

[0280] If the UE 3 is receiving the AI / ML model from the base station 5 via user plane transmission, the UE 3 may receive the AI / ML model via a DRB established between the base station 5 and the UE 3. In this case, the segment may be a PDCP segment with a corresponding PDCP sequence number (SN) or may be generated using a dedicated protocol layer (e.g., the AI / ML protocol layer described above). If the UE 3 is receiving the AI / ML model via a dedicated AI / ML bearer, transmission of the AI / ML model may resume following an RLF. For example, a PDCP status report indicating the status of the AI / ML model transmission (e.g., the SN corresponding to the last received data of the AI / ML model transfer) may be sent by the UE 3 to the base station 5. The base station 5 can then beneficially resume transmission of the AI / ML model based on the SN.

[0281] If the transmission of the AI / ML model to the UE 3 is interrupted due to a handover of the UE 3 from a source base station 5 to a target base station 5 (e.g., as shown in FIG. 5), the UE 3 may be configured to discard segments (or other received data units) of the AI / ML model received from the source base station 5, and the transfer of the model resumes after the handover to the target base station 5 is completed.

[0282] Alternatively, the source base station may notify the UE 3 (e.g., in step S505 of FIG. 5) that it will resume transmitting the AI / ML model following the handover. Following the handover, the UE 3 may provide the target base station 5 with a notification of the status of the transmission of the AI / ML model. For example, the UE may transmit to the target base station 5 a notification of the number of segments (or any other suitable data unit) of the AI / ML model received by the UE 3, a notification of the last segment received by the UE 3, or any other suitable information. In a further alternative, the source base station 5 may already store a notification of the number of segments (or any other suitable data unit) of the AI / ML model transmitted to the UE 3 or a notification of the last segment transmitted to the UE 3 (e.g., in the RRC context information) and can simply transmit the information to the target base station (e.g., via the Xn interface) without first having to receive the information from the UE 3. The source base station 5 may transmit to the target base station information indicating the identity of the AI / ML model that was being transmitted. The source base station 5 may also transmit to the target base station 5 the remaining portions of the AI / ML model to be transmitted to the UE 3. Therefore, advantageously, the target base station 5 can transmit the remaining part of the AI / ML model to the UE 3 even if the AI / ML model has not been previously stored in the target base station 5.

[0283] Alternatively, if the source base station 5 provides notification to the UE 3 that transmission of the AI / ML model will not resume after handover to the target base station, the UE 3 may simply discard (e.g., delete or allow to be overwritten) the portion of the AI / ML model received from the source base station 5.

[0284] Activating and monitoring AI / ML models Next, methods for AI / ML activation and monitoring (e.g., activation tracking) are described. When AI / ML information is broadcast by the network (e.g., the transmission of system information sent by the base station in the manners shown in Figures 18 to 20 in step S1401 of Figures 12 and 13, step S601 of Figure 14), the network may not necessarily have information available indicating whether UE3 is using a particular AI / ML model.

[0285] The network (e.g., base station 5) can perform explicit activation of AI / ML models on UE 3. For example, base station 5 may send an AI / ML model activation request to UE 3, e.g., as shown in step S217 of FIG. 21 . This allows the network to achieve relatively strict control over the AI / ML models used by UE 3 (e.g., if UE 3 is configured not to activate the AI / ML model for use unless UE 3 receives an explicit activation notification from the network). This can be particularly beneficial, for example, when AI / ML is not working as expected and deployment of the model to a relatively small number of UE 3 is desired (e.g., to test the model or to deploy the model only to UE 3 where the model should be deployed). For other AI / ML models, UE 3 can activate the model for use by default (e.g., after receiving the AI / ML model).

[0286] When the base station transmits information indicating the AI / ML functions or AI / ML models to be used by UE3 within the cell of the base station 5 (e.g., transmission of system information transmitted by the base station in the manners shown in Figures 18 to 20 in step S1401 of Figures 12 and 13, step S601 of Figure 14), the base station may include notification of which AI / ML models should be activated by UE3 by default and which AI / ML models require UE3 to receive an activation notification before UE3 can activate the AI / ML models for use.

[0287] Alternatively, the base station 5 may be configured to broadcast AI / ML information regarding AI / ML models that are activated by the UE 3 by default, and to use dedicated signaling (e.g., in step S213 of FIG. 21) to send information regarding AI / ML models that require activation by the network (e.g., notification that the models are available for use).

[0288] Even for AI / ML models that are activated by default by the UE 3, it may be beneficial for the network to obtain information indicating which UEs 3 are using the model. Thus, the UE 3 may be configured to send a notification to the base station 5 that the UE 3 has activated the AI / ML model, even if the AI / ML model was activated by the UE 3 by default (rather than in response to receiving a request to activate the AI / ML model from the network). The notification that the AI / ML model has been activated at the UE 3 may include the AI / ML model ID, the AI / ML model version number, and / or the use case or feature for which the AI / ML model was activated.

[0289] The base station 5 may be configured to send a request to the UE 3 for information about the AI / ML models that the UE 3 is using. After receiving the request, the UE 3 then sends to the base station 5 an indication of the one or more AI / ML models that the UE 3 is using (if any). The UE 3 may send to the base station 5 an AI / ML model ID, an AI / ML model version number, and / or the use case or feature for which the AI / ML model was activated.

[0290] UE capabilities and AI / ML model transfer Further operational considerations for the use of AI / ML models in communication system 1 will now be described. The state of the UE 3 at a particular time may not be suitable for the UE 3 to receive or use the AI / ML model. For example, the UE 3 may not have sufficient memory or storage to receive or use the AI / ML model, or the UE 3 may not have sufficient power or processing resources to execute the AI / ML model.

[0291] The UE 3 may be configured to send a notification to the base station 5 of whether the UE 3 can receive or use (e.g., activate) an AI / ML model. The UE 3 may send the notification to the base station 5 after receiving information from the base station indicating AI / ML models available for use. The UE 3 may provide, for each AI / ML model indicated by the base station 5, a notification of whether the UE 3 can receive and / or use the AI / ML model (e.g., have sufficient memory to store it). The base station 5 can then advantageously determine whether the AI / ML model can be sent to the UE 3 or activated on the UE 3. Alternatively or additionally, the base station 5 may be configured to send a request to the UE 3 for information indicating whether the UE 3 can receive and / or use one or more AI / ML models. Advantageously, for example, if the AI / ML models are stored on the UE 3, the base station 5 can determine whether the UE 3 can activate and use the models. Further alternatively, if the state of the UE 3 does not allow the UE 3 to store and / or use the AI / ML model, the UE 3 may be configured not to initiate acquisition of the AI / ML model (e.g., by requesting transmission of the AI / ML model by a base station).

[0292] User Equipment FIG. 24 is a schematic block diagram showing the main components of a UE 3 for the telecommunications system of FIG. As shown, the UE 3 includes transceiver circuitry 310 operable to transmit signals to and receive signals from a base station 5 via one or more antennas 330 (e.g., comprising one or more antenna elements). The UE 3 includes a controller 370 that controls the operation of the UE 3. The controller 370 is associated with a memory 390 and is connected to the transceiver circuitry 310. Although not necessary for its operation, the UE 3 may, of course, include all the usual functionality of a conventional UE 3 (e.g., a user interface 350, such as a touchscreen / keypad / microphone / speaker, for enabling direct user control and interaction), which may be provided by any one or any combination of hardware, software, and firmware, as appropriate. Software may be pre-installed in the memory 390 and / or downloaded, for example, via a telecommunications network or from a removable data storage device (RMD).

[0293] Controller 370, in this example, is configured to control the overall operation of UE 3 via program or software instructions stored in memory 390. As shown, these software instructions include, among other things, an operating system 410, a communications control module 430, and an AI / ML module 450.

[0294] The communications control module 430 is operable to control communications between the UE 3 and its one or more serving base stations 5 (and other communications devices connected to the base stations 5, such as additional UEs and / or core network nodes). The communications control module 430 is configured to generally handle uplink communications over associated 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 semi-static signaling (e.g., like SRS). The communications control module 430 is also configured to generally handle reception of downlink communications over associated downlink channels (e.g., over the physical downlink control channel (PDCCH) and / or physical downlink shared channel (PDSCH)), including both dynamic and semi-static signaling (e.g., like CSI-RS). The communications control module 430 is responsible for, for example, determining where to monitor downlink control information (e.g., the location of the CSS / USS, CORESET, and associated PDCCH candidates to monitor); determining resources to be used by the UE 3 for transmission / reception of UL / DL communications (including interleaved resources and resources subject to frequency hopping); managing frequency hopping at the UE side; determining how slots / symbols are configured (e.g., for UL, DL, or SBFD communications); determining which bandwidth portion or portions are configured for the UE 3; determining how uplink transmissions should be coded; and appropriately applying any SBFD-specific communications configurations. The communications control module 430 may be configured to control communications (e.g., to send measurement reports according to any of the methods described above) according to any of the methods described above.

[0295] AI / ML module 450 is operable to control the use of AI / ML models in UE 3 (e.g., to generate one or more inferences using the models). AI / ML module 450 may be configured to perform any of the AI / ML-related functions of UE 3 in any of the ways described above.

[0296] base station FIG. 25 is a schematic block diagram illustrating the main components of a base station 5 for the communication system 1 shown in FIG. 1. As shown, the base station 5 includes a transceiver circuit 510 for transmitting signals to and receiving signals from communication devices (such as UE 3) via one or more antennas 530 (e.g., single or multi-panel antenna arrays / large-scale antennas), and a core network interface 550 (e.g., comprising N2, N3, and other reference points / interfaces) for transmitting signals to and receiving signals from network nodes in the core network 7. Although not shown, the base station 5 may also connect to other base stations via appropriate interfaces (e.g., the so-called "Xn" interface in NR). The base station 5 includes a controller 570 that controls the operation of the base station 5. The controller 570 is associated with a memory 590. Software may be pre-installed in the memory 590 and / or downloaded, for example, via the communication system 1 or from a removable data storage device (RMD). The controller 570 is configured, in this example, to control the overall operation of the base station 5 by means of program or software instructions stored in memory 590 .

[0297] As shown, these software instructions include, among other things, an operating system 610, a communications control module 630, and an AI / ML module 650. The communications control module 630 is operable to control communications between the base station 5, the UE 3, and other network entities connected to the base station 5. The communications control module 630 is configured to generally control the reception and decoding of uplink communications over associated 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 semi-static signaling (e.g., SRS). The communications control module 630 is also configured to generally handle the transmission of downlink communications over associated downlink channels (e.g., over the physical downlink control channel (PDCCH) and / or physical downlink shared channel (PDSCH)), including both dynamic and semi-static signaling (e.g., CSI-RS). The communications control module 630 is responsible for managing full-duplex communications (e.g., SBFD), including separation of UL and DL communications over different physical antenna elements, where appropriate. The communication control module 630 is responsible for, for example: determining where the UE 3 should be configured to monitor downlink control information (e.g., the CSS / USS to monitor, the CORESET, and the location of the associated PDCCH candidates); determining resources to be scheduled for the UE transmission / reception of UL / DL communications (including interleaved resources and resources subject to frequency hopping); managing frequency hopping at the base station side; configuring slots / symbols appropriately (e.g., for UL, DL, or SBFD communications, etc.); configuring one or more bandwidth portions for the UE 3; providing related configuration signaling to the UE 3; etc. The communication control module 630 may be configured to control communications (e.g., receive or send mobility information of the UE 3, or a handover request) according to any of the methods described above.

[0298] The AI / ML module 650 may be configured to perform any of the AI / ML-related functions of the UE 3 in any of the ways described above. The base station 5 may be configured to train or retrain the AI / ML models as described above (e.g., in response to UE mobility information fed back to the base station 5 from another node in the network, such as another base station 5).

[0299] Core Network Nodes / Functions Figure 26 is a block diagram illustrating the main components of a core network node or function, such as the AMF, CPF, UPF, SMF, or OAM. As shown, the core network function includes a transceiver circuit 710 operable to transmit signals to and receive signals from other nodes (including UEs 3, base stations 5, and other core network nodes) via a network interface 720. A controller 730 controls the operation of the core network function in accordance with software stored in memory 740. The software may be pre-installed in memory 740 and / or may be downloaded, for example, via the 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.

[0300] The communication control module 760 is responsible for handling (generating / sending / receiving) signaling between core network functions and other nodes such as the UE 3, base station 5, and other core network nodes. The signaling may include, for example, UE context / UE capability notifications for the UE 3 related to energy saving.

[0301] As shown in Figure 21, 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 any of the methods described above. The core network node / function may be configured to train or retrain the AI / ML model as described above (e.g., in response to UE mobility information being fed back to the core network node / function from another node in the network, such as base station 5).

[0302] Modifications and Alternatives As those skilled in the art will appreciate, multiple modifications and alternatives can be made to the above embodiments while still benefiting from the disclosure embodied therein. Although the above examples are described with reference to AI / ML models, it will be appreciated that the above methods are advantageous even when the model is not an AI / ML model. Any other suitable type of model or function can be used to generate an inference (e.g., a decision or prediction).

[0303] For example, terms specific to a cellular communication generation (e.g., 2G, 3G, 4G, 5G, 6G, etc.) may be used to refer to a particular communication entity for clarity, but it will be understood that technical features described for a given entity are not limited to devices of that particular communication generation. The technical features may be implemented in any functionally equivalent communication entity regardless of the terms used to refer to them.

[0304] In the above description, the UE and base station have been described for ease of understanding as having several separate functional components or modules. While these modules may be provided in this manner in certain applications, for example, where an existing system is modified to implement the present disclosure, in other applications, for example, in systems designed from the beginning with the features of the present invention in mind, these modules may be incorporated into the overall operating system or code and therefore may not be identifiable as separate entities.

[0305] In the above exemplary embodiments, several software modules have been described. As will be appreciated by those skilled in the art, the software modules may be provided in compiled or non-compiled form, and may be supplied as a signal over a computer network or on a recording medium. Furthermore, 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 updating the base station or UE to update its functionality.

[0306] Each controller may comprise any suitable form of processing circuitry, including, for example, but not limited to, one or more hardware-implemented computer processors, microprocessors, central processing units (CPUs), arithmetic logic units (ALUs), input / output (IO) circuitry, internal memory / cache (program and / or data), processing registers, communication buses (such as, for example, a control bus, a data bus, and / or an address bus), direct memory access (DMA) facilities, hardware or software-implemented counters, pointers, and / or timers, etc. Various other modifications will be apparent to those skilled in the art and will not be described in further detail herein.

[0307] The base station may comprise a "distributed" base station having a central unit "CU" and one or more individual distributed units (DUs). User equipment (or "UE," "mobile station," "mobile device," or "wireless device") in this disclosure is an entity that connects to a network via a wireless interface.

[0308] It should be noted that the present disclosure is not limited to dedicated communication devices, but can be applied to any device having communication capabilities as described in the following paragraphs.

[0309] The terms "user equipment" or "UE" (as this term is used by 3GPP), "mobile station," "mobile device," and "wireless 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. It will be understood that the terms "mobile station" and "mobile device" also encompass devices that remain stationary for extended periods of time.

[0310] The UE may be, for example, an item of equipment for production or manufacturing and / or an item of energy-related machinery (such as, for example, equipment or machinery such as 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 application systems thereof, tools, dies or molds, rolls, conveying equipment, elevators, material handling equipment, textile machinery, sewing machinery, printing and / or related machinery, paper converting machinery, chemical machinery, mining machinery and / or construction machinery and / or related equipment, machinery and / or implements for the agricultural, forestry, and / or fisheries industries, 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 foregoing equipment or machinery, etc.).

[0311] A UE may be, for example, an item of transportation equipment (e.g., transportation equipment such as rail cars, automobiles, motorcycles, bicycles, trains, buses, carts, rickshaws, ships and other watercraft, aircraft, rockets, satellites, drones, balloons, etc.) A UE may be, for example, an item of information and communications equipment (e.g., information and communications equipment such as electronic computers and related equipment, communications and related equipment, electronic components, etc.).

[0312] The UE may be, for example, a refrigerator, a refrigerator application product, an item of trade and / or service industry equipment, a vending machine, an automated service machine, an office machine or equipment, a home appliance or electronic device (e.g., household appliances such as audio equipment, video equipment, loudspeakers, radios, televisions, microwave ovens, rice cookers, coffee machines, dishwashers, washing machines, dryers, electric fans or related equipment, vacuum cleaners, etc.).

[0313] The UE may be, for example, an electrical application system or device (such as, for example, an electrical application system or device, such as an x-ray system, a particle accelerator, a radioisotope device, a sonic device, an electromagnetic application device, a power application device, etc.).

[0314] The UE may be, for example, an electronic lamp, a lighting fixture, a measuring instrument, an analyzer, a tester, or a surveying or detecting device (such as, for example, a smoke alarm, a human alarm sensor, a motion sensor, a radio tag, or other surveying or detecting device), a watch or clock, laboratory equipment, optical equipment, medical equipment and / or systems, a weapon, an edged item, a hand tool, etc.

[0315] The UE may be, for example, a wirelessly equipped personal digital assistant or related equipment (such as a wireless card or module designed to be attached to or inserted into another electronic device (e.g., a personal computer, an electrical measuring machine, etc.)). The UE may be part of a device or system that uses various wired and / or wireless communication technologies to provide the applications, services, and solutions described below with respect to the "internet of things (IoT)."

[0316] Internet of Things devices (or "Things") may be equipped with appropriate electronics, software, sensors, network connections, etc. that enable these devices to collect and exchange data with each other and other communicating devices. IoT devices may comprise automated machines that follow software instructions stored in internal memory. IoT devices may operate without the need for human supervision or interaction. IoT devices may also remain stationary and / or inactive for extended periods of time. IoT devices may be implemented as part of (typically) stationary equipment. IoT devices may also be incorporated into non-stationary equipment (e.g., vehicles) or attached to animals or people being monitored / tracked.

[0317] It will be appreciated that IoT technologies may be implemented on any communication device that can connect to a communication network to send / receive data, whether such communication device is controlled by human input or by software instructions stored in memory.

[0318] It will be appreciated that IoT devices may also be referred to as Machine-Type Communication (MTC) devices or Machine-to-Machine (M2M) communication devices. It will be appreciated 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 some examples of machine-type communication applications. [Table 2] [Table 3]

[0319] The applications, services, and solutions may include MVNO (Mobile Virtual Network Operator) services, emergency wireless 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 wireless 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 NW selection services, function restriction services, PoC (Proof of Concept) services, personal information management services, ad hoc networks / DTN (Delay Tolerant Networking) services, etc.

[0320] Furthermore, the above-mentioned UE categories are merely examples of applications of the concepts and exemplary embodiments described herein, and it should be understood that these concepts and embodiments are not limited to the above-mentioned UEs and may be subject to various modifications.

[0321] Various other modifications will be apparent to those skilled in the art and will not be described in further detail here.

[0322] All or part of the exemplary embodiments disclosed above can be described as follows, but are not limited to the following: (Appendix 1) 1. A method performed by an access network node, the method comprising: transmitting to a user equipment (UE) a notification of a feature to be implemented in a cell of the access network node, the feature being implemented using a corresponding model for determining, predicting, or generating output parameters; receiving a request for a model from a UE; model transmission information including an indication of one or more communications resources to be used by the UE to receive the model; or Model to the UE; and A method comprising: (Appendix 2) 2. The method of claim 1, wherein the model is an artificial intelligence or machine learning, AI / ML, model. (Appendix 3) The method includes transmitting, to a UE, information indicating one or more communication resources used by the UE to request model information corresponding to the model; receiving a request for model information from a UE; sending model information to a UE; 3. The method of claim 1 or 2, further comprising: (Appendix 4) 4. The method of claim 3, wherein the model information includes at least one of an indication of the identity of the model or a version number of the model. (Appendix 5) The method comprises: receiving information from the UE indicative of a characteristic of the UE; determining at least one of a model to send to the UE or a configuration of the model to send to the UE based on a characteristic of the UE; 5. The method of any one of claims 1 to 4, further comprising: (Appendix 6) 6. The method of claim 5, wherein the UE characteristics include at least one of a capability of the UE, a type of the UE, an indication of a model supported by the UE, or an indication of a configuration of a model supported by the UE. (Appendix 7) 7. The method of claim 5 or 6, further comprising sending to the UE at least one of an indication of the model to be sent to the UE, an indication of a configuration of the model to be sent to the UE, or a size of the model. (Appendix 8) 8. The method of any one of claims 5 to 7, wherein the model configuration includes one or more parameters for use with the model to generate a determination, prediction, or output parameter. (Appendix 9) 9. The method of any one of claims 1 to 8, further comprising transmitting the model to the UE using the indicated communication resources. (Appendix 10) 9. A method according to any one of Supplementary Notes 1 to 8, wherein the signalled communications resources are for use by the UE to receive models from a node other than an access network node. (Appendix 11) 11. The method of claim 10, wherein the node other than the access network node is a server that stores the model or a core network node. (Appendix 12) 12. The method of claim 10 or 11, wherein the notified communication resources include at least one of a network address of a node other than an access network node, or a configuration of a radio bearer for receiving a model from a node other than an access network node. (Appendix 13) The method comprises: receiving a request for a model in a radio resource control (RRC) transmission; transmitting the model to the UE using RRC transmission; 13. The method of any one of appendices 1 to 12, comprising: (Appendix 14) the model is an AI / ML model, and the method comprises: receiving a request for a model using a dedicated protocol layer for the transmission of information related to the AI / ML model; or Transmitting the model to the UE using a dedicated protocol layer 13. The method of any one of claims 1 to 12, comprising at least one of: (Appendix 15) the model is an AI / ML model, and the method comprises: receiving a request for a model in an RRC transmission; transmitting the model to the UE using a dedicated protocol layer for transmission of information related to the AI / ML model; 14. The method of any one of appendices 1 to 13, comprising: (Appendix 16) 1. A method performed by an access network node, the method comprising: From user equipment (UE), Features supported by the UE, which are implemented using a corresponding model for generating decision, prediction, or output parameters; a model for generating decision, prediction, or output parameters supported by the UE; or Notification of one or more models stored in the UE receiving UE capability information indicating at least one of: determining a model to be sent to or activated in the UE based on the UE capability information; model transmission information including an indication of one or more communications resources to be used by the UE to receive the model; Model, or UE request to activate the model to the UE; and A method comprising: (Appendix 17) The method comprises: determining, based on the UE capability information, to request further UE capability information from the UE; sending a request for further UE capability information to the UE; receiving further UE capability information from the UE; determining a model to be sent to the UE based on the further UE capability information; 17. The method of claim 16, further comprising: (Appendix 18) Further UE capability information is Notification of the model versions supported by the UE, or Notification of one or more models stored in the UE 18. The method of claim 17, comprising at least one of: (Appendix 19) 19. The method of any one of Supplementary Notes 16 to 18, wherein the signaled one or more communications resources are for use by the UE to receive a model from a node other than an access network node. (Appendix 20) 20. The method of claim 19, wherein the node other than the access network node is a server that stores the model or a core network node. (Appendix 21) 21. The method of claim 19 or 20, wherein the notified communication resources include at least one of a network address of a node other than an access network node, or a configuration of a radio bearer for receiving a model from a node other than an access network node. (Appendix 22) 22. The method of any one of Supplementary Notes 16 to 21, wherein the method comprises, after sending the model to the UE, the UE sending a request for the UE to activate the model. (Appendix 23) 22. The method of any one of claims 16 to 21, wherein the method comprises determining that a model is stored in the UE; and sending a request by the UE to activate the model stored in the UE. (Appendix 24) 1. A method performed by a user equipment (UE), the method comprising: receiving, from an access network node, notification of features to implement within a cell of the access network node, the features being implemented using a corresponding model for generating determination, prediction, or output parameters; determining to obtain a model; sending a request for the model to an access network node; model transmission information including an indication of one or more communications resources to be used by the UE to receive the model; or Model from the access network node; A method comprising: (Appendix 25) 25. The method of claim 24, wherein determining to acquire the model includes determining to acquire the model if the model is not stored in the UE. (Appendix 26) The method includes receiving, from an access network node, information indicating one or more communications resources to be used by the UE to request model information corresponding to the model; sending a request for model information to an access network node; receiving model information from an access network node; 26. The method of claim 24 or 25, further comprising: (Appendix 27) 27. The method of claim 26, wherein the model information includes at least one of an indication of the identity of the model and a version number of the model. (Appendix 28) 28. The method of any one of claims 24 to 27, further comprising receiving a model from an access network node using the indicated communication resources. (Appendix 29) 29. The method of any one of Supplementary Notes 24 to 28, wherein the signaled communications resources are for use by the UE to receive a model from a node other than an access network node, and the method comprises receiving the model from a node other than an access network node. (Appendix 30) 30. The method of claim 29, wherein the node other than the access network node is a server that stores the model or a core network node. (Appendix 31) 31. The method of claim 29 or 30, wherein the notified communication resources include at least one of a network address of a node other than an access network node, or a configuration of a radio bearer for receiving a model from a node other than an access network node. (Appendix 32) 32. A method according to any one of claims 29 to 31, wherein the method comprises sending a request for the model to a node other than the access network node based on the notified communications resources. (Appendix 33) 1. A method performed by a user equipment (UE), the method comprising: In the access network node, Features supported by the UE, which are implemented using a corresponding model for generating decision, prediction, or output parameters; a model for generating decision, prediction, or output parameters supported by the UE; or Notification of one or more models stored in the UE transmitting UE capability information indicating at least one of: model transmission information including an indication of one or more communications resources to be used by the UE to receive the model; Model, or UE request to activate the model from the access network node; A method comprising: (Appendix 34) 34. The method of claim 33, wherein the UE activates the model when the UE receives a request for the UE to activate the model. (Appendix 35) The method comprises: receiving a request for further UE capability information from an access network node; transmitting further UE capability information to the access network node; 35. The method of claim 33 or 34, further comprising: (Appendix 36) Further UE capability information is Notification of the model versions supported by the UE, or Notification of one or more models stored in the UE 36. The method of claim 35, comprising at least one of: (Appendix 37) 37. The method of any one of Supplementary Notes 33 to 36, wherein the signaled one or more communications resources are for use by the UE to receive a model from a node other than an access network node. (Appendix 38) 38. The method of claim 37, wherein the node other than the access network node is a server that stores the model or a core network node. (Appendix 39) 39. The method of claim 37 or 38, wherein the notified communication resources include at least one of a network address of a node other than an access network node, or a configuration of a radio bearer for receiving a model from a node other than an access network node. (Appendix 40) 40. The method of any one of Supplementary Notes 33 to 39, wherein if the UE receives the model from the access network node, the method further comprises sending to the access network node a notification of whether the model has been received at the UE. (Appendix 41) 1. A method performed by a user equipment (UE), the method comprising: a UE receives, from an access network node, a portion of a model for generating decisions, predictions, or output parameters using a wireless link between the UE and the access network node, and a wireless link failure occurs after the portion of the model has been received from the access network node but before the remaining portion of the model has been received by the UE; maintaining a portion of the model in a memory of the UE; re-establishing a radio link between the UE and the access network node or establishing a radio link between the UE and another access network node; When a radio link is re-established between the UE and the access network node, sending to an access network node a notification of a portion of the model stored in the UE; receiving the remainder of the model from the access network node; When a radio link is established with another access network node, sending a notification of the part of the model stored in the UE to another access network node; receiving the remainder of the model from another access network node; A method comprising: (Appendix 42) If portions of the model are received at the UE in multiple data transfer units, each data transfer unit contains a sub-portion of the model; the notification of the portion of the model stored at the UE includes notification of the identity of the last data transfer unit received at the UE; 41. The method described in Appendix 41. (Appendix 43) 43. The method of claim 42, wherein the notification of the identity of the last data transfer unit received by the UE comprises notification of the identity of a radio resource control (RRC) segment, or notification of the identity of a packet data convergence protocol (PDCP) sequence number (SN). (Appendix 44) If the wireless link is established with another access network, the method further comprises: To another access network node, Indication of the model's identity, or The part of the model received by the UE is the notification of the identity of the access network node Send at least one of 44. The method of any one of claims 41 to 43, further comprising: (Appendix 45) 44. The method of any one of claims 41 to 43, wherein when the radio link is re-established with the access network node, the method further comprises sending a notification of the identity of the model to the access network node. (Appendix 46) 1. A method performed by a user equipment (UE), the method comprising: receiving from a first access network node a portion of a model for generating decisions, predictions, or output parameters; performing a handover procedure for handing over the UE from the first access network node to the second access network node; maintaining a portion of the model in the memory of the UE3 during the handover procedure; sending a notification to a second access network node of the portion of the model stored in the UE; receiving the remainder of the model from another access network node; A method comprising: (Appendix 47) 47. The method of claim 46, further comprising receiving notification from the first access network node or the second access network node that the UE will receive the remainder of the model from the second access network node. (Appendix 48) portions of the model are received at the UE from the first access network node in a plurality of data forwarding units, each data forwarding unit including a sub-portion of the model; the notification of the portion of the model stored at the UE includes notification of the identity of the last data transfer unit received at the UE; 48. The method according to claim 46 or 47. (Appendix 49) The method of claim 48, wherein the notification of the identity of the last data transfer unit received by the UE includes notification of the identity of a radio resource control (RRC) segment, or notification of the identity of a packet data convergence protocol (PDCP) sequence number or SN. (Appendix 50) 1. A method performed by an access network node, the method comprising: an access network node transmits to a user equipment, UE, a portion of a model for generating decision, prediction, or output parameters using a wireless link between the UE and the access network node, and a wireless link failure occurs after the portion of the model is transmitted to the UE but before the remaining portion of the model is transmitted to the UE; re-establishing a radio link between the UE and an access network node; receiving notification from the UE of portions of the model stored in the UE; determining a remaining portion of the model to be transmitted to the UE based on the notification of the portion of the model stored in the UE; transmitting the remainder of the model to the UE; and A method comprising: (Appendix 51) portions of the model are transmitted to the UE in a plurality of data transfer units, each data transfer unit containing a sub-portion of the model; the notification of the portion of the model stored at the UE includes notification of the identity of the last data transfer unit received at the UE; 51. The method described in Appendix 50. (Appendix 52) 52. The method of claim 51, wherein the notification of the identity of the last data transfer unit received by the UE comprises notification of the identity of a radio resource control (RRC) segment, or notification of the identity of a packet data convergence protocol (PDCP) sequence number (SN). (Appendix 53) The UE is in radio resource control,RRC,connected state when the access network node sends the model part to the UE; The access network node maintains the RRC connection state and associated context after a radio link failure occurs; 53. The method of any one of appendices 50 to 52. (Appendix 54) 1. A method performed by a first access network node, the method comprising: transmitting to a user equipment (UE) a portion of the model for generating decision, prediction, or output parameters; sending a notification to the UE that the remainder of the model should be received from a second access network node; performing a handover procedure for handing over the UE from the first access network node to the second access network node; A method comprising: (Appendix 55) 55. The method of claim 54, further comprising transmitting the remaining portion of the model to the second access network node for transmission of the remaining portion of the model from the second access network node to the UE. (Appendix 56) 56. The method of claim 54 or 55, further comprising sending a notification of the identity of the model to the second access network node. (Appendix 57) A method performed by a second access network node, the method comprising: performing a handover procedure for handing over the UE from the first access network node to the second access network node; receiving, from the UE or from a first access network node, notification of a portion of a model for generating decision, prediction, or output parameters to be stored in the UE or notification of a remaining portion of the model to be sent to the UE; transmitting the remainder of the model to the UE; and A method comprising: (Appendix 58) 58. The method of claim 57, further comprising receiving a remaining portion of the model from the first access network node. (Appendix 59) 59. The method of claim 57 or 58, further comprising receiving notification of the identity of the model from the first access network node. (Appendix 60) 1. A method performed by a user equipment (UE), the method comprising: receiving a model for generating decision, prediction, or output parameters; determining to activate a model for use in the UE; activating the model for use in the UE; determining to send a notification to an access network node that the model has been activated for use in the UE; sending a notification to an access network node that the model has been activated for use in the UE; A method comprising: (Appendix 61) 61. The method of claim 60, wherein determining to send a notification that the model has been activated for use in the UE includes determining to send a notification to an access network node when the model is received in a broadcast transmission. (Appendix 62) 1. A method performed by a user equipment (UE), the method comprising: receiving a model for generating decision, prediction, or output parameters; receiving notification from an access network node that a model is to be activated for use in the UE; determining, based on the notification, to activate the model for use in the UE; activating the model for use in the UE; A method comprising: (Appendix 63) 63. The method of claim 62, wherein the model is an artificial intelligence or machine learning, AI / ML, model. (Appendix 64) 1. A method for a user equipment (UE), comprising: receiving notification from an access network node of models for generating decision, prediction, or output parameters supported for use in a cell of the access network node; sending an indication to an access network node that the UE is able to receive and use the model; A method comprising: (Appendix 65) 64. The method according to claim 64, further comprising: receiving a request from an access network node for notification of whether the UE can receive and use the model; and the UE sending notification of whether it can receive and use the model to the access network node after receiving the request. (Appendix 66) 66. The method of claim 64 or 65, wherein the indication of whether the UE can receive and use the model includes an indication of at least one of a state of memory resources in the UE, a state of processing resources in the UE, or a state of power resources in the UE. (Appendix 67) 1. A method of an access network node, the method comprising: sending to a user equipment, UE, notification of models for generating decision, prediction, or output parameters supported for use in a cell of the access network node; receiving, from the UE, an indication that the UE is able to receive or use the model; determining, based on the received notification, whether the model should be sent to the UE or activated for use at the UE; A method comprising: (Appendix 68) 68. The method of claim 67, wherein the indication of whether the UE can receive and use the model includes an indication of at least one of a state of memory resources at the UE, a state of processing resources at the UE, or a state of power resources at the UE. (Appendix 69) an access network node, means for transmitting to a user equipment (UE) notification of features to be implemented in a cell of the access network node, the features being implemented using a corresponding model for determining, predicting, or generating output parameters; means for receiving a request for a model from a UE; Including, The means of transmission is model transmission information including an indication of one or more communications resources to be used by the UE to receive the model; or Model and configured to transmit to the UE at least one of Access network node. (Appendix 70) an access network node, From user equipment (UE), Features supported by the UE, which are implemented using a corresponding model for generating decision, prediction, or output parameters; a model for generating decision, prediction, or output parameters supported by the UE; or Notification of one or more models stored in the UE means for receiving UE capability information indicating at least one of: means for determining a model to be sent to or activated in the UE based on the UE capability information; model transmission information including an indication of one or more communications resources to be used by the UE to receive the model; Model, or UE request to activate the model and means for transmitting at least one of the above to the UE. (Appendix 71) A user equipment, UE, means for receiving, from an access network node, notification of features to implement in a cell of the access network node, the features being implemented using a corresponding model for generating determination, prediction, or output parameters; a means for determining whether to acquire a model; means for transmitting a request for a model to an access network node; The means of receiving model transmission information including an indication of one or more communications resources to be used by the UE to receive the model; or Model configured to receive from the access network node at least one of UE. (Appendix 72) A user equipment, UE, In the access network node, Features supported by the UE, which are implemented using a corresponding model for generating decision, prediction, or output parameters; a model for generating decision, prediction, or output parameters supported by the UE; or Notification of one or more models stored in the UE means for transmitting UE capability information indicating at least one of: model transmission information including an indication of one or more communications resources to be used by the UE to receive the model; Model, or UE request to activate the model from an access network node; Including, UE. (Appendix 73) a user equipment, UE, wherein the UE receives from an access network node a portion of a model for generating decisions, predictions, or output parameters using a wireless link between the UE and the access network node, and a wireless link failure occurs after the portion of the model has been received from the access network node but before the remaining portion of the model has been received by the UE; Discarding parts of the model, sending a request for the model when a radio link between the UE and the access network node is re-established; receiving a model from an access network node; Configured for,UE. (Appendix 74) a user equipment, UE, wherein the UE receives from an access network node a portion of a model for generating decisions, predictions, or output parameters using a wireless link between the UE and the access network node, and a wireless link failure occurs after the portion of the model has been received from the access network node but before the remaining portion of the model has been received by the UE; Maintaining a portion of the model in UE3 memory, re-establishing a radio link between the UE and the access network node or establishing a radio link between the UE and another access network node; When a radio link is re-established between the UE and the access network node, sending to an access network node a notification of a portion of the model stored in the UE; receiving the remainder of the model from the access network node; When a radio link is established with another access network node, sending a notification of the part of the model stored in the UE to another access network node; receiving the remainder of the model from another access network node; Configured for,UE. (Appendix 75) A user equipment, UE, means for receiving from a first access network node a portion of a model for generating decisions, predictions, or output parameters; means for performing a handover procedure for handing over the UE from a first access network node to a second access network node; means for maintaining a portion of the model in the memory of the UE3 during the handover procedure; means for sending to a second access network node a notification of the part of the model stored in the UE; means for receiving the remainder of the model from another access network node; Including, UE. (Appendix 76) an access network node, the access network node transmitting to a user equipment, UE, a portion of a model for generating decision, prediction, or output parameters using a wireless link between the UE and the access network node, and a wireless link failure occurring after the portion of the model has been transmitted to the UE but before the remainder of the model has been transmitted to the UE; re-establishing a radio link between the UE and the access network node; and receiving notification from the UE of portions of the model stored in the UE; determining a remaining portion of the model to be transmitted to the UE based on the notification of the portion of the model stored in the UE; transmitting the remainder of the model to the UE; and An access network node configured for (Appendix 77) a first access network node, transmitting to a user equipment (UE) a portion of the model for generating decision, prediction, or output parameters; and sending a notification to the UE that the remainder of the model should be received from the second access network node; a transmitting means configured for means for performing a handover procedure for handing over the UE from a first access network node to a second access network node; a first access network node including: (Appendix 78) a second access network node, means for performing a handover procedure for handing over the UE from a first access network node to a second access network node; means for receiving, from the UE or from a first access network node, notification of a portion of a model for generating decision, prediction or output parameters that is stored in the UE or notification of a remaining portion of the model that is to be sent to the UE; means for transmitting the remainder of the model to the UE; a second access network node including: (Appendix 79) A user equipment, UE, means for receiving a model for generating decision, prediction, or output parameters; means for determining to activate a model for use in the UE; means for activating the model for use in the UE; means for determining to send a notification to an access network node that the model has been activated for use in the UE; means for sending a notification to an access network node that the model has been activated for use in the UE; Including, UE. (Appendix 80) A user equipment, UE, receiving a model for generating decision, prediction, or output parameters; receiving notification from an access network node that a model is to be activated for use in the UE; receiving means configured for means for determining, based on the notification, to activate the model for use in the UE; means for activating the model for use in the UE; Including, UE. (Appendix 81) A user equipment, UE, means for receiving, from an access network node, notification of models for generating decision, prediction or output parameters supported for use in a cell of the access network node; means for sending to an access network node an indication that the UE is able to receive and use the model; Including, UE. (Appendix 82) an access network node, means for transmitting to a user equipment, UE, notification of models for generating decision, prediction, or output parameters supported for use in a cell of the access network node; means for receiving, from the UE, an indication that the UE is able to receive or use the model; means for determining, based on the received notification, whether the model should be sent to the UE or activated for use in the UE; an access network node, including:

[0323] This application is based on and claims the benefit of priority from UK Patent Application No. 2302236.1, filed February 16, 2023, the disclosure of which is incorporated herein by reference in its entirety. [Explanation of symbols]

[0324] 3. User Equipment 5 (Radio)Access Network ((R)AN) nodes, base stations, gNB 5-1 First base station 5-2 Second base station 7 Core Network 9 cells 10 control plane function (CPF) 10-1 Access management function 10-2 Session management function 10-n Other Features 11 User Plane Function (UPF) 41 Data Collection Function 43 Model training function 45 Model inference function 47 Actors 50 Distributed Units (DU) 60 Central Unit (CU) 151 AI / ML Server 310 Transceiver Circuit 330 Antenna 350 User Interface 390 memory 410 Operating System 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 Interface 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 System 630 Communication Control Module 650 AI / ML Module 710 Transceiver Circuit 720 network interface 730 Controller 740 memory 750 Operating Systems 760 Communication Control Module 770 AI / ML Module

Claims

1. 1. A method performed by a user equipment (UE), the method comprising: receiving information from an access network node indicating an operating condition related to artificial intelligence or machine learning (AI / ML); applying the operating conditions for the AI / ML model based on the information; and A method comprising:

2. The operating conditions are: Information indicating the version of the AI / ML model executed by the UE; Information indicating usage of the AI / ML model executed by the UE; information indicating that explicit activation by the access network node is required; and Indicate at least one of The method of claim 1.

3. sending information to the access network node indicating at least one AI / ML model that the UE is running; The method of claim 1 or 2, further comprising:

4. said transmitting being performed upon request from said access network node; The method of claim 4.

5. The operating conditions are: whether the UE should implement the AI / ML model; whether the UE should activate the AI / ML model, or whether the UE should receive the AI / ML model from the access network node; Indicate at least one of The method of claim 1.

6. The operating conditions are: whether the UE is authorized to run the AI / ML model; whether the UE is authorized to activate the AI / ML model, or whether the UE is authorized to receive the AI / ML model; and wherein the UE status information is used to determine at least one of the following: The method according to claim 1 or 5.

7. The status information is Information indicating buffer availability for the UE; or information indicating a power level of the UE; at least one of The method of claim 6.

8. from the access network node: Identity information of the AI / ML model; The size of the AI / ML model; receiving at least one of determining that the UE should initiate a procedure for receiving the AI / ML model; The method of claim 1 , further comprising:

9. If the UE should initiate the UE procedure to receive the AI / ML model, sending a message to the access network node to initiate the procedure for the UE to receive the AI / ML model; receiving the AI / ML model from the access network node; The method of claim 8 further comprising:

10. receiving common information related to the AI / ML model from the access network node, the common information comprising: Application of said AI / ML model; UE vendor, The model of the UE, or Version of the AI / ML model receiving, which is common to at least one of: sending to the access network node a request message to receive the identity information of the AI / ML model, the request message comprising: UE capabilities with respect to said AI / ML model; Information indicating the vendor of the UE; Information indicating the model of the UE; or Information indicating the version of the AI / ML model that the UE has; transmitting the information indicating at least one of: Further comprising: receiving identity information of the AI / ML model based on the request message; 10. The method according to claim 8 or 9.

11. said access network node; At least one usage of at least one AI / ML model supported by the UE; or At least one AI / ML model supported by the UE. transmitting UE capability information indicating at least one of: receiving a request message from the access network node to transmit UE capabilities related to the AI / ML model; sending detailed information to the access network node indicating the UE capabilities with respect to the AI / ML model; Further comprising: The receiving of the identity information of the AI / ML model is performed based on the UE capability for the AI / ML model.

10. The method according to claim 8 or 9.

12. receiving the AI / ML model; Radio Resource Control (RRC) layer, New layers used for AI / ML model transmission, Application layer, At least one layer lower than the RRC layer; or a data radio bearer between the UE and the access network node; Implemented using at least one of:

12. The method according to any one of claims 9 to 11.

13. receiving the AI / ML model; after receiving at least one signaling from the access network node, the signaling including assistance information for the receiving of the AI / ML model; and Using the support information, To be carried out, 13. The method according to any one of claims 9 to 12.

14. The support information is a resource configuration used by a UE to receive the AI / ML model; a radio bearer configuration used by the UE to receive the AI / ML model; or Scheduling to receive the AI / ML model Indicate at least one of The method of claim 13.

15. receiving the AI / ML model; at least one dedicated transmission; at least one multicast transmission, or At least one broadcast transmission performed by at least one of 15. The method according to any one of claims 9 to 14.

16. Transmitting a status of said receiving said AI / ML model. Further comprising: The status of receiving the AI / ML model is: The receiving of the AI / ML model was successfully completed, or Is it necessary to resubmit the AI / ML model? Indicate at least one of 16. The method according to any one of claims 9 to 15.

17. said message being received by said access network node; external entities, a core network node, or Service Management Orchestration Server forwarded to at least one of The AI / ML model is the external entity, the core network node; or said service management orchestration server from said at least one of transmitted via the access network node; 17. The method according to any one of claims 9 to 16.

18. If the receiving of the AI / ML model is interrupted, Discarding at least one segment of the AI / ML model received by the UE; restarting said receiving said AI / ML model; 18. The method of any one of claims 9 to 17, further comprising:

19. If the receiving of the AI / ML model is interrupted, the same cell as before the receiving of the AI / ML model was interrupted, or Different cells resuming said receiving said AI / ML model on 18. The method of any one of claims 9 to 17, further comprising:

20. the different cell has the same identity of the AI / ML model as the cell that was used before the receiving of the AI / ML model was interrupted.

20. The method of claim 19.

21. The information indicating at least one segment of the AI / ML model received by the UE includes: stored in the UE's context, transmitted to the access network node; the AI / ML model is transmitted to a further access network node operating a cell used for the resuming of receiving the AI / ML model; 21. The method of claim 19 or 20.

22. the further access network node is different from the access network node; the information indicating the at least one segment of the AI / ML model received by the UE is transmitted from the UE to the access network node; handover related information is transmitted from said access network node to said further access network node; The handover related information is information identifying the AI / ML model; information indicating the at least one segment of the AI / ML model transmitted by the access network node; or At least one remaining segment of the AI / ML model at least one of 22. The method of claim 21.

23. receiving the AI / ML model; Radio link failure, or Handover is interrupted due to 23. The method of any one of claims 18 to 22.

24. 1. A method performed by an access network node, said method comprising: Transmitting information to a user equipment (UE) indicating operating conditions associated with an artificial intelligence or machine learning (AI / ML) model for use by the UE in applying the operating conditions associated with the AI / ML model. A method comprising:

25. A user equipment, means for receiving information from an access network node indicative of an operating condition related to artificial intelligence or machine learning (AI / ML); means for applying the operating conditions for the AI / ML model based on the information; [0023] A user equipment including:

26. an access network node, means for transmitting, to a user equipment (UE), information indicating operating conditions associated with an artificial intelligence or machine learning (AI / ML) model for use by the UE in applying the operating conditions associated with the AI / ML model; an access network node, including:

Images