Methods and apparatus for AI / ML data transfer

Abstract

A first entity transmits artificial intelligence / machine learning (AI / ML) data to a second entity using multicast-broadcast services (MBS) or a local area data network (LAN), based on information or s

Description

BACKGROUND Field

[0001] Certain examples of the present disclosure relate to methods, apparatus and / or systems for transferring an Artificial Intelligence (Al) I Machine Learning (ML) model and / or associated information. Further, certain examples of the present disclosure relate to methods and apparatus for transferring an AI / ML model and / or AI / ML associated information to one or more devices using multicast / broadcast services (MBS). Further, certain examples of the present disclosure relate to methods and apparatus for providing an architecture to support AI / ML data transfer via MBS. Additionally, certain examples of the present disclosure relate to methods and apparatus for using local area data network (LADN) protocol data unit (PDU) sessions to transfer AI / ML model and / or associated information. Description of Related Art

[0002] The content of the following documents is referred to below and / or their content provides background information that the following disclosure should be considered in the context of: [1] 3GPP TS 22.261 - Service requirements for the 5G system, SA1, Release 18 (e.g., V18.7.0). [2] 3GPP TS 23.247 - 5G; Architectural enhancements for 5G multicast-broadcast services, Release 17 (e.g., V17.4.0). [3] 3GPP TS 38.300 - 5G; NR; NR and NG-RAN Overall description; Stage-2, Release 17 (e.g., V17.2.0). (Note: the example versions shown for each TS are non-limiting, other versions of the TS may be considered also)

[0003] Wireless or mobile (cellular) communications networks in which a mobile terminal (e.g., user equipment (UE), such as a mobile handset) communicates via a radio link with a network of base stations, or other wireless access points or nodes, have undergone rapid development through a number of generations. The 3rd Generation Partnership Project (3GPP) design, specify and standardise technologies for mobile wireless communication networks. Fourth Generation (4G) and Fifth Generation (5G) systems are now widely deployed.

[0004] 3GPP standards for 4G systems include an Evolved Packet Core (EPC) and an Enhanced-UTRAN (E-UTRAN: an Enhanced Universal Terrestrial Radio Access Network). The E-UTRAN uses Long Term Evolution (LTE) radio technology. LTE is commonly used to refer to the whole system including both the EPC and the E-UTRAN, and LTE is used in this sense in the remainder of this document. LTE should also be taken to include LTE enhancements such as LTE Advanced and LTE Pro, which offer enhanced data rates compared to LTE.

[0005] In 5G systems a new air interface has been developed, which may be referred to as 5G New Radio (5G NR) or simply NR. NR is designed to support the wide variety of services and use case scenarios envisaged for 5G networks, though builds upon established LTE technologies. New frameworks and architectures are also being developed as part of 5G networks in order to increase the range of functionality and use cases available through 5G networks. One such new framework is the use of artificial intelligence I machine learning (AI / ML), which may be used for the optimisation of the operation of 5G networks.

[0006] In AI / ML operation, AI / ML models and / or data might be transferred across the AI / ML applications (e.g., application functions (AFs)), 5GC (5G core), UEs (user equipments) etc.). Without limitation, the AI / ML works could be divided into two main phases: model training and inference. During model training and inference, multiple rounds of interaction may be required.

[0007] In Section 6.40 (‘AI / ML model transfer in 5GS’) in TS 22.261 [1], three types of AI / ML operations to be supported in Release 18 are described as follows: a) AI / ML operation splitting between AI / ML endpoints The AI / ML operation / model is split into multiple parts according to the current task and environment. The intention is to offload the computation-intensive, energy-intensive parts to network endpoints, whereas leave the privacy-sensitive and delay-sensitive parts at the end device. The device executes the operation / model up to a specific part / layer and then sends the intermediate data to the network endpoint. The network endpoint executes the remaining parts / layers and feeds the inference results back to the device. b) AI / ML model / data distribution and sharing over 5G system Multi-functional mobile terminals might need to switch the AI / ML model in response to task and environment variations. The condition of adaptive model selection is that the models to be selected are available for the mobile device. However, given the fact that the AI / ML models are becoming increasingly diverse, and with the limited storage resource in a UE, it can be determined to not pre-load all candidate AI / ML models on-board. Online model distribution (i.e. new model downloading) is needed, in which an AI / ML model can be distributed from a NW (network) endpoint to the devices when they need it to adapt to the changed AI / ML tasks and environments. For this purpose, the model performance at the UE needs to be monitored constantly. c) Distributed / Federated Learning over 5G system The cloud server trains a global model by aggregating local models partially-trained by each end devices. Within each training iteration, a UE performs the training based on the model downloaded from the Al server using the local training data. Then the UE reports the interim training results to the cloud server via 5G UL channels. The server aggregates the interim training results from the UEs and updates the global model. The updated global model is then distributed back to the UEs and the UEs can perform the training for the next iteration.

[0008] 3GPP agreed Rel-18 Study on Artificial Intelligence (Al) / Machine Learning (ML) for NR Air Interface (referring to 3GPP technical specification group (TSG) RAN meeting #94-3, RP-213599). As part of 3GPP TSG RAN1 work on this Study, RAN1 concluded the following assumptions and agreements: RAN1#109: Take the following network-UE collaboration levels as one aspect for defining collaboration levels 1. Level x: No collaboration 2. Level y: Signaling-based collaboration without model transfer 3. Level z: Signaling-based collaboration with model transfer Note: Other aspect(s), for defining collaboration levels is not precluded and will be discussed in later meetings, e.g., with / without model updating, to support training / inference, for defining collaboration levels will be discussed in later meetings FFS: Clarification is needed for Level x-y boundary RAN1#110bis Working Assumption Include the following into a working list of terminologies to be used for RAN1 AI / ML air interface SI discussion. Terminology Description AI / ML model delivery A generic term referring to delivery of an AI / ML model from one entity to another entity in any manner. Note: An entity could mean a network node / function (e.g., gNB, LMF, etc.), UE, proprietary server, etc. RAN1#110bis Working Assumption • Define Level y-z boundary based on whether model delivery is transparent to 3gpp signalling over the air interface or not. • Note: other procedures than model transfer / delivery are decoupled with collaboration level y-z • Clarifying note: Level y includes cases without model delivery. RAN2#119bis > R2 assumes that for the existing (under discussion) AI / ML use cases, proprietary models may be supported and / or open format may be supported. > R2 assumes that from Management or Control point of view mainly some meta info about a model may need to be known, details FFS. > R2 assumes that a model is identified by a model ID. Its usage is FFS. > General FFS: AIML Model delivery to the UE may have different options, Controlplane (multiple subvariants), User Plane, can be discussed case by case. (FFS: for further study)

[0009] AI / ML model delivery from the network to a device / devices (e.g., a UE / UEs) may depend on the assumed collaboration level between the network and the device. For example, for network-UE collaboration level z (referring to the above: ‘Signaling-based collaboration with model transfer’), the network would need to transfer the model (fully or partially) and any model associated information to the UE, using Control plane (i.e. RRC signalling or NAS signalling) or User Plane options.

[0010] However, considering that the AI / ML model size could be large, depending for example on model use case, functionality, scenario, or configuration and / or model structure / format, and / or model input and output data set size, then transferring the model and any associated model information may result in high increase in signalling overhead and radio interface resources. This may be especially so if the network needs to deliver the model to more than one UE at a given time and / or location.

[0011] The use of control plane (CP) becomes inefficient due to a large number of UEs and / or a large amount of data to be shared - this may overwhelm the signalling plane of the system and is hence inefficient.

[0012] The use of user plane (UP), as defined currently, via the establishment of a PDU session towards a data network may be inefficient since the data network may not be topologically close to the UE(s) with which data need to be shared with minimal latency.

[0013] As such, the common CP or UP methods may not be efficient for AI / ML especially when the data needs to be shared as quickly as possible with minimal (or at least reduced) latency, where the latency if high may negatively impact the AI / ML application and / or objective. SUMMARY

[0014] It is an aim of certain examples of the present disclosure to address, solve and / or mitigate, at least partly, at least one of the problems and / or disadvantages associated with the related art, for example at least one of the problems and / or disadvantages described herein. It is an aim of certain examples of the present disclosure to provide at least one advantage over the related art, for example at least one of the advantages described herein.

[0015] According to an aspect of the present disclosure, there is provided a first entity included in a communications network, the first entity comprising: a transmitter; a receiver; and a controller configured to: receive, from a second entity included in the communications network, first information or signalling associated with transfer of artificial intelligence / machine learning (AI / ML) data; and transmit first AI / ML data to the second entity using multicast-broadcast services (MBS) or a local area data network (LADN), based on the first information or signalling.

[0016] According to an example, the first information or signalling comprises an indication that the second entity supports MBS.

[0017] According to another example, the controller is configured to: based on receiving the indication, transmit, to the second entity, second information or signalling for configuring the second entity to request the first AI / ML data; and receive, from the second entity, a request for the first AI / ML data; and the first AI / ML data is transmitted using MBS in response to the request.

[0018] According to another example, the first information or signalling comprises a request for the first AI / ML data; and the first AI / ML data is transmitted to the second entity in response to the request.

[0019] According to another example, the request comprises information identifying one or more AI / ML models and / or information associated with one or more AI / ML models; and the first AI / ML data comprises the identified one or more AI / ML models and / or information associated with one or more AI / ML models.

[0020] According to another example, the controller is configured to: transmit, to the second entity, one or more parameters for receiving the first AI / ML data using MBS; and / or transmit, to the second entity, an indication of presence of MBS data for a AI / ML service; and wherein the first AI / ML data is transmitted to the second entity using MBS.

[0021] According to another example, the one or more parameters are transmitted using dedicated signalling or system information, or signalling using the application layer; and / or the one or more parameters comprises one or more of: at least one Temporary Mobile Group Identity (TMGI), at least one Internet Protocol (IP) address, or information on an area of multicast.

[0022] According to another example, the first entity is a network function (NF) included in a 5G core (5GC) connected to a radio access network (RAN), and is connected to one or more MBS-related function; and the controller is configured to: collect data for AI / ML related to the RAN, the collected data including the first AI / ML data; and provide the first AI / ML data into a MBS framework for transmission to the second entity.

[0023] According to another example, the controller is configured to: collect data regarding delivery modes of previous MBS traffic and / or status of the RAN; and provide, to the connected one or more MBS-related function, a recommendation on a delivery mode decision for the second entity and / or a recommendation to switch a delivery mode.

[0024] According to another example, a protocol data unit (PDU) session is established between the second entity and the first entity for transfer of the AI / ML data.

[0025] According to another example, the controller is configured to: deploy the LADN in an area for transferring the AI / ML data to the second entity; and transmit the first AI / ML data to the second entity using the LADN when a protocol data unit (PDU) session is established with the second entity.

[0026] According to another example, the first information or signalling comprises an indication that the second entity supports LADN AI / ML connection or a request for LADN information; and the controller is configured to: in response to receiving the first information or signalling, transmit the LADN information to the second entity, where the LADN information is for use in establishing the PDU session or the LADN information is requested for transfer of the first AI / ML data.

[0027] According to another example, wherein the LADN information comprises a list of tracking area identifiers (TAIs) indicating where the PDU session can be obtained.

[0028] According to another aspect of the present disclosure, there is provided a second entity included in a communications network, the second entity comprising: a transmitter; a receiver; and a controller configured to: transmit, to a first entity included in the communications network, first information or signalling associated with transfer of artificial intelligence / machine learning (AI / ML) data; and receive first AI / ML data to the second entity using multicast-broadcast services (MBS) or a local area data network (LADN), based on the first information or signalling.

[0029] According to another example, the first information or signalling comprises an indication that the second entity supports MBS.

[0030] According to another example, the controller is configured to: receive, from the first entity, second information or signalling for configuring the second entity to request the first AI / ML data; and transmit, to the first entity, a request for the first AI / ML data based on the second information or signalling; and wherein the first AI / ML data is received using MBS.

[0031] According to another example, the first information or signalling comprises a request for the first AI / ML data; and wherein the first AI / ML data is received using MBS.

[0032] According to another example, the request comprises information identifying one or more AI / ML models and / or information associated with one or more AI / ML models; and the first AI / ML data comprises the identified one or more AI / ML models and / or information associated with one or more AI / ML models.

[0033] According to another example, the controller is configured to: receive, from the first entity, one or more parameters for receiving the first AI / ML data using MBS; and / or receive, from the first entity, an indication of presence of MBS data for a AI / ML service; and the first AI / ML data is received from the first entity using MBS based on the one or more parameters and / or the indication of presence.

[0034] According to another example, the one or more parameters are transmitted using dedicated signalling or system information, or signalling using the application layer; and / or the one or more parameters comprises one or more of: at least one Temporary Mobile Group Identity (TMGI), at least one Internet Protocol (IP) address, or information on an area of multicast.

[0035] According to another example, the controller is configured to join an MBS session for receiving the first AI / ML data based on one or more of: determining the second entity is in an area where MBS is available for transfer of AI / ML data, based on the one or more parameters and / or the indication of presence; determining a current location of the second entity is within an area indicated in the indication of presence; receiving the indication of presence in response to a request, transmitted from the second entity to the first entity, for the first AI / ML data; or determining a current location of the second entity is within a satellite coverage area where the MBS is available, based on satellite ephemeris information, the one or more parameters and / or the indication of presence.

[0036] According to another example, the controller is configured to: establish a protocol data unit (PDU) session with the first entity for receiving the first AI / ML data using MBS, based on one or more MBS parameters received from the first entity or from a third entity in the communications network.

[0037] According to another example, the first information or signalling comprises an indication that the second entity supports LADN AI / ML connection or a request for LADN information; and wherein the controller is configured to: receive the LADN information from the first entity, where the LADN information is for use in establishing a LADN protocol data unit (PDU) session with the first entity or the LADN information is requested for transfer of the first AI / ML data.

[0038] According to another example, the LADN information comprises a list of tracking area identifiers (TAIs) indicating where the LADN PDU session can be obtained; the controller is configured to establish the LAD PDU session based on determining a current location of the second entity is within a TAI included in the list of TAIs; and the first AI / ML data is received via the LADN PDU session.

[0039] According to another example (e.g. relating to any of the above examples or aspects), the first entity is one of 5G core (5GC), radio access network (RAN) or a network function (NF); and / or the second entity is a user equipment (UE) or a plurality of UEs.

[0040] According to another aspect of the present disclosure, there is provided a method of a first entity included in a communications network, the method comprising: receiving, from a second entity included in the communications network, first information or signalling associated with transfer of artificial intelligence / machine learning (AI / ML) data; and transmitting first AI / ML data to the second entity using multicast-broadcast services (MBS) or a local area data network (LADN), based on the first information or signalling.

[0041] According to another aspect of the present disclosure, there is provided a method of a second entity included in a communications network, the method comprising: transmitting, to a first entity included in the communications network, first information or signalling associated with transfer of artificial intelligence / machine learning (AI / ML) data; and receiving first AI / ML data to the second entity using multicast-broadcast services (MBS) or a local area data network (LADN), based on the first information or signalling.

[0042] Accordingly another aspect of the present disclosure, there is provided a network comprising: a first entity according to any aspect or example identified above, and a second entity according to any aspect or example identified above.

[0043] It will be appreciated that all combinations of the above examples and aspects are envisaged and should be considered to be included herein. Additionally, for each example relating to a first entity, a second entity or a network comprising a first entity and a second entity, there should also be seen to be provided a corresponding example of a method by / of / for the first entity, a method by / of / for a second entity, or a method by / of / for the network, respectively.

[0044] According to another aspect of the present disclosure, there is provided a computer readable storage medium comprising instructions which, when executed by one or more processor of an electronic device, cause the electronic device to perform a method according to any of the aspects and / or examples identified above.

[0045] Other aspects, advantages, and salient features of the invention will become apparent to those skilled in the art from the following detailed description taken in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Embodiments / examples of the present disclosure are further described hereinafter with reference to the accompanying drawings, in which: Figure 1 is “Figure 4.1-1: Delivery Methods” from TS 23.247 version 17.4.0 [2], Figure 2 is “Figure 5.1-1: 5G System architecture for Multicast and Broadcast Service” from TS 23.247 version 17.4.0 [2], Figure 3 is “Figure 5.1-2: 5G System architecture for Multicast and Broadcast Service in reference point representation" from TS 23.247 version 17.4.0 [2], Figure 4 illustrates a system architecture in accordance with examples of the present disclosure. Figure 5 is a block diagram illustrating an example structure of a network entity in accordance with certain examples of the present disclosure. DETAILED DESCRIPTION

[0047] The following description of examples of the present disclosure, with reference to the accompanying drawings, is provided to assist in a comprehensive understanding of certain examples of the present invention. The description includes various specific details to assist in that understanding but these are to be regarded as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the examples described herein can be made without departing from the scope of the invention or disclosure.

[0048] The same or similar components may be designated by the same or similar reference numerals, although they may be illustrated in different drawings.

[0049] Detailed descriptions of techniques, structures, constructions, functions or processes known in the art may be omitted for clarity and conciseness, and to avoid obscuring the subject matter of the present disclosure.

[0050] The terms and words used herein are not limited to the bibliographical or standard meanings, but are merely used to enable a clear and consistent understanding of the invention.

[0051] Throughout the description of this specification, the words “comprise”, “include” and “contain” and variations of the words, for example “comprising” and “comprises”, means “including but not limited to”, and is not intended to (and does not) exclude other features, elements, components, integers, steps, processes, operations, functions, characteristics, properties and / or groups thereof.

[0052] Throughout the description of this specification, the singular form, for example “a”, “an” and “the”, encompasses the plural unless the context otherwise requires. For example, reference to “an object” includes reference to one or more of such objects.

[0053] Throughout the description, the expression “at least one of A, B and / or C” (or the like) and the expression “one or more of A, B and / or C” (or the like) should be seen to separately include all possible combinations, for example: A, B, C, A and B, A and C, A and B and C.

[0054] Throughout the description of this specification, language in the general form of “X for Y” (where Y is some action, process, operation, function, activity or step and X is some means for carrying out that action, process, operation, function, activity or step) encompasses means X adapted, configured or arranged specifically, but not necessarily exclusively, to do Y.

[0055] Features, elements, components, integers, steps, processes, operations, functions, characteristics, properties and / or groups thereof described or disclosed in conjunction with a particular aspect, embodiment or example are to be understood to be applicable to any other aspect, embodiment or example described herein unless incompatible therewith..

[0056] Certain examples of the present disclosure relate to methods, apparatus and / or systems etc. for transferring an Artificial Intelligence (Al) I Machine Learning (ML) model and / or associated information. Further, certain examples of the present disclosure relate to methods and apparatus for transferring an AI / ML model and / or AI / ML associated information to one or more devices using multicast / broadcast services (MBS). Further, certain examples of the present disclosure relate to methods and apparatus for providing an architecture to support AI / ML data transfer via MBS. Additionally, certain examples of the present disclosure relate to methods and apparatus for using local area data network (LADN) protocol data unit (PDU) sessions to transfer AI / ML model and / or associated information. Note, however, that the present disclosure is not limited to these examples, and includes other examples.

[0057] The following examples are applicable to, and use terminology associated with, 3GPP 5G. However, the skilled person will appreciate that the techniques disclosed herein are not limited to these examples or to 3GPP 5G, and may be applied in any suitable system or standard, for example one or more existing and / or future generation wireless communication systems or standards. The skilled person will appreciate that the techniques disclosed herein may be applied in any existing or future releases of 3GPP 5G NR or any other relevant standard. For example, the functionality of the various network entities and other features disclosed herein may be applied to corresponding or equivalent entities or features in other communication systems or standards. Corresponding or equivalent entities or features may be regarded as entities or features that perform the same or similar role, function, operation or purpose within the network.

[0058] A particular network entity may be implemented as a network element on a dedicated hardware, as a software instance running on a dedicated hardware, and / or as a virtualised function instantiated on an appropriate platform, e.g. on a cloud infrastructure.

[0059] The skilled person will appreciate that the present invention is not limited to the specific examples disclosed herein. For example: • The techniques disclosed herein are not limited to 3GPP 5G. • One or more entities in the examples disclosed herein may be replaced with one or more alternative entities performing equivalent or corresponding functions, processes or operations. • One or more of the messages in the examples disclosed herein may be replaced with one or more alternative messages, signals or other type of information carriers that communicate equivalent or corresponding information. • One or more further elements, entities and / or messages may be added to the examples disclosed herein. • One or more non-essential elements, entities and / or messages may be omitted in certain examples. • The functions, processes or operations of a particular entity in one example may be divided between two or more separate entities in an alternative example. • The functions, processes or operations of two or more separate entities in one example may be performed by a single entity in an alternative example. • Information carried by a particular message in one example may be carried by two or more separate messages in an alternative example. • Information carried by two or more separate messages in one example may be carried by a single message in an alternative example. • The order in which operations are performed may be modified, if possible, in alternative examples. • The transmission of information between network entities is not limited to the specific form, type and / or order of messages described in relation to the examples disclosed herein.

[0060] Certain examples of the present disclosure may be provided in the form of an apparatus / device / network entity configured to perform one or more defined network functions and / or a method therefor. Such an apparatus / device / network entity may comprise one or more elements, for example one or more of receivers, transmitters, transceivers, processors, controllers, modules, units, and the like, each element configured to perform one or more corresponding processes, operations and / or method steps for implementing the techniques described herein. For example, an operation / function of X may be performed by a module configured to perform X (or an X-module). Certain examples of the present disclosure may be provided in the form of a system (e.g., a network) comprising one or more such apparatuses / devices / network entities, and / or a method therefor.

[0061] It will be appreciated that examples of the present disclosure may be realized in the form of hardware, software or a combination of hardware and software. Certain examples of the present disclosure may provide a computer program comprising instructions or code which, when executed, implement a method, system and / or apparatus in accordance with any aspect, example and / or embodiment disclosed herein. Certain embodiments of the present disclosure provide a machine-readable storage storing such a program.

[0062] As discussed above, the common CP or UP methods may not be efficient for use with AI / ML especially when the data needs to be shared as quickly as possible with minimal (or at least reduced) latency, where the latency if high may negatively impact the AI / ML application and / or objective. Certain examples of the present invention provide methods, systems, apparatus etc. which aim to address this problem and / or related issues.

[0063] Certain examples of the present disclosure provide methods, systems, apparatus etc. for efficiently sharing AI / ML data with a (potentially large) set of devices (e.g., UEs). In particular, certain embodiments include the delivering (or transmitting, providing, sending etc.) of AI / ML model(s) to device(s) based on multicast / broadcast service (MBS).

[0064] In TS 23.247 [2], an overview of multicast and broadcast communication includes the following details (note that the definition of various abbreviations is given in the Annex below): 4.1 Overview of multicast and broadcast communication Between 5GC and NG-RAN, there are two possible delivery methods to transmit the MBS data: - 5GC Individual MBS traffic delivery method: This method is only applied for multicast MBS sessions. 5GC receives a single copy of MBS data packets and delivers separate copies of those MBS data packets to individual UEs via per-UE PDU sessions, hence for each such UE one PDU session is required to be associated with a Multicast MBS session. - 5GC Shared MBS traffic delivery method: This method is applied for both broadcast and multicast MBS sessions. 5GC receives a single copy of MBS data packets and delivers a single copy of those MBS packets to an NG-RAN node, which then delivers the packets to one or multiple UEs. The 5GC Shared MBS traffic delivery method is required in all MBS deployments. The 5GC Individual MBS traffic delivery method is required to enable mobility when there is an NG-RAN deployment with non-homogeneous support of MBS. For the Multicast MBS session, a single copy of MBS data packets received by the CN may be delivered via 5GC Individual MBS traffic delivery method for some UE(s) and via 5GC Shared MBS traffic delivery method for other UEs. Between the NG-RAN and the UE, two delivery methods are available for the transmission of MBS data packets over radio interface: - Point-to-Point (PTP) delivery method: NG-RAN delivers separate copies of MBS data packets over radio interface to individual UE(s). - Point-to-Multipoint (PTM) delivery method: NG-RAN delivers a single copy of MBS data packets over radio interface to multiple UEs. NG-RAN may use a combination of PTP / PTM to deliver an MBS data packets to UEs. NOTE 2: The PTP and PTM delivery methods are defined in RAN WGs. As depicted in the following figure, 5GC Shared MBS traffic delivery method (with PTP or PTM delivery) and 5GC Individual MBS traffic delivery method may be used at the same time for a multicast MBS session.

[0065] The figure referred to in the final paragraph above is shown in Figure 1, which corresponds to Fig. 4.1-1 (‘Delivery Methods’) from TS 23.247 [2] as indicated previously.

[0066] TS 23.247 also discloses the architecture of Multicast and Broadcast Service, as follows and with reference to Figure 2 and Figure 3. Note that Fig. 2 corresponds to Fig. 5.1-1 (“5G System architecture for Multicast and Broadcast Service”) from TS 23.247 [2] as indicated previously, and Fig. 3 corresponds to Fig. 5.1-2 (“5G System architecture for Multicast and Broadcast Service in reference point representation”) from TS 23.247 [2] as indicated previously; accordingly, in the following excerpt of TS 23.247 [2], references to “Figure 5.1-1” may be understood with reference to Fig. 2 and references to “Figure 5.1-2” may be understood with reference to Fig. 3. From TS 23.247: 5.1 General architecture Figure 5.1-1 depicts the MBS reference architecture. Service-based interfaces are used within the Control Plane. Support for interworking at reference points xMB and MB2 is described in Annex C. NOTE 1: The MBSF is optional and may be collocated with the NEF or AF / AS, and the MBSTF is an optional network function. NOTE 2: The existing service-based interfaces of Nnrf, Nudm, and Nsmf are enhanced to support MBS. The existing service-based interfaces of Npcf and Nnef are enhanced to support MBS. NOTE 3: A MBS-enabled AF uses either Nmbsf or Nnef to interact with the MBSF. Figure 5.1-2 depicts the 5G system architecture for MBS using the reference point representation. NOTE 4: The existing reference points of N1, N2, N4, N5, N10, N11, N30 and N33 are enhanced to support MBS. 5 NOTE 5: Regarding the functionalities, Nmb13, N29mb and Nmb1 are identical, Nmb5 and Nmb10 are identical, Nmb9 and N6mb are identical.

[0067] Regarding the above, it should be noted that the actual data to be shared for the MBS service is provided by the AF using the Nmb8 reference point. Furthermore, the AF is assumed to be an external entity which interacts with the 3GPP system 10 via the NEF (Network Exposure Function).

[0068] In TS 38.300 [3], various disclosures relating to Multicast and Broadcast Services are as follows: 16.10 Multicast and Broadcast Services 16.10.1 General 15 NR system enables resource efficient delivery of multicast / broadcast services (MBS). For broadcast communication service, the same service and the same specific content data are provided simultaneously to all UEs in a geographical area (i.e., all UEs in the broadcast service area as defined in TS 23.247 are authorized to receive the data). A 20 broadcast communication service is delivered to the UEs using a broadcast session. A UE can receive a broadcast communication service in RRC_IDLE, RRC_INACTIVE and RRC_CONNECTED state. For multicast communication service, the same service and the same specific content 25 data are provided simultaneously to a dedicated set of UEs (i.e., not all UEs in the multicast service area as defined in TS 23.247 are authorized to receive the data). A multicast communication service is delivered to the UEs using a multicast session. A UE can receive a multicast communication service in RRC_CONNECTED state with mechanisms such as PTP and / or PTM delivery, as defined in clause 16.10.5.4. HARQ 30 feedback / retransmission can be applied to both PTP and PTM transmission. 16.10.5 Multicast Handling 16.10.5.2 Configuration 35 A UE can receive data of MBS multicast session only in RRC_CONNECTED state. If the UE which joined a multicast session is in RRC_CONNECTED state and when the multicast session is activated, the gNB sends RRCReconfiguration message with relevant MBS configuration for the multicast session to the UE. When there is (temporarily) no data to be sent to the UEs for a multicast session, the gNB 40 may move the UE to RRC IDLE / INACTIVE state. gNBs supporting MBS use a group notification mechanism to notify the UEs in RRC IDLE / INACTIVE state when a multicast session has been activated by the CN or the gNB has multicast session data to deliver. Upon reception of the group notification, the UEs reconnect to the network. The group notification is addressed with P-RNTI on PDCCH, and the paging channels are monitored by the UE as described in clause 9.2.5. Paging message for group notification contains MBS session ID which is utilized to page all UEs in RRC IDLE and RRC INACTIVE states that joined the associated MBS multicast session, i.e., UEs are not paged individually. The UE stops monitoring for group notifications related to a specific multicast session once the UE leaves this multicast session. If the UE in RRC IDLE state that joined an MBS multicast session is camping on gNB not supporting MBS, the UE may be notified about multicast session activation or data availability by CN-initiated paging where CN pages each UE individually, as described in clause 9.2.5. If the UE in RRC INACTIVE state that joined MBS multicast session is camping on gNB not supporting MBS, the UE may be notified about data availability individually by RAN-initiated paging, as described in clause 9.2.5.

[0069] In 3GPP RP-213568 “New WID: Enhancements of NR Multicast and Broadcast Services” (Rel-18), the following is disclosed: This Work Item is to further enhance the NR Multicast / Broadcast functions based on Rel-17 MBS. The objectives for Rel-18 include: - Specify support of multicast reception by UEs in RRC_INACTIVE state [RAN2, RAN3] o PTM configuration for UEs receiving multicast in RRC_INACTIVE state [RAN2] o Study the impact of mobility and state transition for UEs receiving multicast in RRC_INACTIVE. (Seamless / lossless mobility is not required) [RAN2, RAN3] - Specify Uu signalling enhancements to allow a UE to use shared processing for MBS broadcast and unicast reception, i.e., including UE capability and related assistance information reporting regarding simultaneous unicast reception in RRC_CONNECTED and MBS broadcast reception from the same or different operators [RAN2] - Study and if necessary, specify enhancements to improve the resource efficiency for MBS reception in RAN sharing scenarios [RAN3] Note: collaboration with SA2 is expected in due course for the above objectives.

[0070] The above passages of TS 23.247, TS 38.300 and RP-213568 are provided to assist in further understanding of certain examples / embodiments according to the present disclosure.

[0071] According to various embodiments of the present disclosure, there is provided a system, method or apparatus etc. for delivering a AI / ML model(s) (and / or associated information) to devices or network entities based on MBS architecture. Certain embodiments may reduce signalling overhead resulting from model delivery to device(s) by benefiting from the efficient usage of network resources of the MBS systems. Certain embodiments may also improve on the latency as the data may be close to the network (e.g., radio access network (RAN)) and hence requires less time to be transported to the device.

[0072] Certain embodiments of the present disclosure relate to AI / ML model transfer from the network (e.g., one or more network entities (e.g., a network node, network function, virtual entity, logical entity etc.)) to the device(s). A non-limiting example of a device is a UE; in the below, where there is mention of a UE, it will be appreciated that the present disclosure is not limited to a UE but may refer to a terminal or another device (e.g., a generic device) instead - the reference to a UE is merely to more-clearly illustrate examples of the present disclosure.

[0073] Certain examples of the present disclosure include the network (for example, RAN node, CN entity, and / or any other network entity) delivering / transferring to the UE(s) an AI / ML model using multicast-broadcast services. In an embodiment, a network (e.g., any one or more network entities) is configured to deliver or transfer an AI / ML model(s) to one or more devices using MBS.

[0074] In another embodiment, the network (e.g. RAN node, CN entity, and / or any other network entity) delivers / transfers to the UE(s) an AI / ML model and any associated model information using multicast-broadcast services. In an another example, the network delivers to the UE(s) information associated with an AI / ML model using multicast-broadcast services.

[0075] In an embodiment, the network (e.g. RAN node, CN entity, and / or any other network entity) may configure the device(s), with capability to support multicastbroadcast services, to request the transfer / delivery of an AI / ML model and / or any associated model information via multicast-broadcast services. For example, the network may transmit information or signalling to the device(s) to configure the device(s) to request the transfer / delivery of an AI / ML model and / or any associated model information via MBS.

[0076] In another example, the network (e.g. RAN node, CN entity, and / or any other network entity) may deliver an AI / ML model and / or any associated model information, to the UE(s) located in a given area via multicast-broadcast services. For example, the network may identify location information or a region to which an AI / ML model(s) is to be delivered, and deliver the model to one or more UE(s) located in the location / region.

[0077] In an embodiment, the UE indicates to the network (e.g. a network entity, as per the examples above) whether it supports delivery / transfer of an AI / ML model and / or any associated model information via multicast-broadcast services. For example the UE may transmit data, information, a signal etc. to the network (e.g., to a RAN node, CN entity, gNB and / or any network entity) to inform the network that delivery / transfer of an AI / ML model and / or any associated model information via multicast-broadcast services is supported by the UE.

[0078] In an another embodiment, the UE may request the network to deliver / transfer more than one AI / ML model and / or associated model(s) information. For example the UE may transmit information identifying one or more AI / ML models and / or associated model(s) information to be delivered to the UE.

[0079] In the aforementioned examples / embodiments, the existing MBS framework may be reused and the network may feed the AI / ML data (e.g., AI / ML model(s), AI / ML model associated information, AI / ML related information etc.) into the MBS framework. The device (e.g., UE) may be provided (e.g., by the network or one or more network entities) with the relevant parameters to receive this data, e.g., Temporary Mobile Group Identity (TMGI) or Internet Protocol (IP) addresses (e.g. Source specific IP multicast address for IPv4, or Source specific IP multicast address for IPv6), area of multicast (e.g. service area, list of TAIs, etc).

[0080] In certain examples, the TMGI, and / or any other parameter that is required by the UE to receive this data, may be configured in the UE or provided to the UE by the network, e.g., using dedicated signaling (NAS and / or RRC signaling / messages) or system information (e.g. broadcast periodically or on-demand) or signaling using the application layer. For example, a UE (or other device) may attempt to join an MBS session when any one (or more) of the following occurs: • The UE is in an area where MBS service is available for AI / ML, where the UE may make this determination. For example, the UE may determine, based on the current tracking area identity (TAI) and the knowledge of where an MBS session is available in an area or a TAI. • The UE receives (e.g., from the network) an indication of presence of MBS data for AI / ML in the current UE location (e.g. current TAI or MBS service area, etc). This indication may be received via RRC or NAS signaling or system information. • The UE receives (e.g., from the network) an indication of availability or presence of MBS data for AI / ML following a previous UE request for this data. This indication may be received via RRC or NAS signaling or system information, while the UE request may be provided using RRC or NAS signaling. • The UE is in a satellite coverage area where MBS service is available for AI / ML, where the UE may make this determination. For example, the UE may determine it is in a satellite coverage area based on knowledge of satellite ephemeris or other information.

[0081] Certain examples of the present disclosure relate to new architecture to support AI / ML data transfer via MBS.

[0082] In an embodiment, a network function (NF) is defined to be within the 5G core (5GC) which may be connected to the RAN (e.g. more than one radio network entity such as gNB) and, optionally, may also connect to any one or more of a / the Multicast / Broadcast Service Function (MBSF), Multicast / Broadcast Service Transport Function (MBSTF), Multicast / Broadcast Session Management Function (MB-SMF) and Multicast / Broadcast User Plane Function (MB-UPF).

[0083] In a non-limiting example, the NF (which may be a new NF) may be referred to as an AI / ML NF or RAN AI / ML NF. This terminology is used at times below to assist in illustrating features of the present disclosure, however the present disclosure should not be seen as limited thereto and the term NF (or even ‘network entity’) may be considered in place of AI / ML NF and RAN AI / ML NF (or the like).

[0084] The NF is configured to communicate with any network entity such as but not limited to the RAN, AMF, SMF, UPF, UDM, PCF, NEF, etc, or any of the entities listed above and / or below.

[0085] According to certain examples of the present disclosure, there is provided the architecture shown in Figure 4, where a new entity 100 (AI / ML NF 100) and connections 111, 113, 115, 117, 119 to other network entities are shown (for instance, compare with Fig. 2). It will be appreciated that the AI / ML NF 100 may be connected to other nodes / entities as described above, although this is not shown in the figure for brevity. Additionally, it will be appreciated that the AI / ML NF 100 may be connected to fewer nodes and / or entities than illustrated in the figure -essentially the AI / ML NF 100 may be connected to any of the illustrated network entities as desired and as appropriate, and also to other, non-illustrated network entities.

[0086] In certain examples, one function of the NF 100 (i.e., AI / ML NF 100) may be to collect data for AI / ML which is related to the RAN and, optionally, process this data and, optionally, feed it (e.g., deliver, transmit etc.) into the MBS framework for sharing with the UE(s). As such, the NF 100 may take any one or more of the actions and / or responsibilities of the AF / AS node (see Fig. 4) and hence may mimic, at least to some extent, the AF / AS node. However this NF 100 may be local to the 5GC and the NF 100 may feed the AI / ML data to the MBSTF, or any other MBS node, so that it is shared to UEs with reduced latency; where the reduced latency may arise, for example, because this node (AI / ML NF 100) may be residing closer to the RAN whilst also connected to the MBS framework.

[0087] In certain examples, a UE may establish, with this AI / ML NF 100, a PDU session, where a specific Data Network Name (DNN) and / or slice (e.g., Single Network Slice Selection Assistance Information (S-NSSAI)) may be reserved for this purpose, such that the RAN AI / ML NF 100 may appear to be the endpoint AF / AS for AI / ML application. As such, in a further example, the NF 100 may also share the TMGI or any other MBS parameter with the UE, where the parameter(s) may be used for joining a session to receive AI / ML data.

[0088] In other examples, the MBS parameter(s) may be provided by the 5GC to the UE; e.g., the TMGI may be provided by the RAN (using any RRC message and / or information element (IE), which may be new or existing, or using system information, e.g., new or existing system information block(s) (SIB(s))), or by the Access and Mobility Management Function (AMF) (for example, using any existing or new Non-Access Stratum (NAS) message and / or IE), or by the Session Management Function (SMF) (for example, using any existing or new NAS message and / or IE), or by any other entity. For example, the 5GC may behave as described as in the examples above (i.e., to provide the MBS parameter(s) to the UE) when the subscription information requires so. Note that the 5GC entities may share this information amongst themselves. For example, the AMF may obtain this indication from the subscription information and either share the information with the UE using any NAS message, or may provide it to the RAN (for example, via NG interface signalling, for instance, part of the UE context procedures, INITIAL CONTEXT SETUP REQUEST message and / or UE CONTEXT MODIFICATION REQUEST message, or AMF CP RELOCATION INDICATION message, UE INFORMATION TRANSFER message , HANDOVER REQUEST message and / or PATH SWITCH REQUEST ACKNOWLEDGE message) where the RAN may in turn provide it to the UE using any RRC message (or system information).

[0089] In certain embodiments, the 5GC may behave as described above optionally for a UE which indicates capability to perform AI / ML (e.g. for the RAN). For instance, receiving an indication that the UE has capability to perform AI / ML (e.g., model training, inference etc.) triggers the 5GC to behave in accordance with one of the examples, embodiments, aspects etc. described above / herein.

[0090] Further to the above, once the UE receives the information (that is, a / any MBS parameter for AI / ML (which may also be preconfigured in the UE)), the UE may establish a new PDU session for this purpose (e.g., for AI / ML), or may join an MBS session for AI / ML (optionally, when the UE is in the MBS service area e.g., based on the TAI), and receive the AI / ML data associated with a TMGI of interest.

[0091] In certain examples, further to the above, the UE may locally save the received data and use it for AI / ML accordingly.

[0092] In another embodiment, the 5GC may leverage the AI / ML NF 100 to determine the best, or optimal, traffic delivery mode for the AI / ML traffic over the 5GC, namely 5GC Individual MBS traffic delivery or 5GC Shared MBS traffic delivery. As explained herein, these delivery modes differ in whether copies of each traffic packets are individually transmitted to each UE, or just a single copy is sent to RAN and shared with all UEs. The AI / ML NF 100 may collect data regarding the delivery modes of past MBS traffic as well as the status on the network and provide a recommendation to the MBS system (e.g. MBSF and / or MBSTF and / or MB-SMF and / or MB-UPF) on the delivery mode decision for one UE, a group of UEs, or all UEs. Additionally or alternatively, the AI / ML NF 100 may also indicate to the MBS system (e.g. MBSF and / or MBSTF and / or MB-SMF and / or MB-UPF) that a delivery mode switch is recommended (e.g. 5GC Individual MBS traffic delivery to 5GC Shared MBS traffic delivery or vice versa).

[0093] Certain examples of the present disclosure relate to using Local Area Data Network (LADN) PDU sessions to transfer AI / ML model (and / or associated AI / ML model information).

[0094] In an embodiment, the network (e.g., any network entity such as one of the nonlimiting examples described above / herein) deploys LADN in an area which can then transfer AI / ML to the UE when the UE establishes a PDU session for LADN.

[0095] In certain examples, the UE may explicitly request LADN information for AI / ML, where the request may contain an explicit indication (e.g., a new indication) that the LADN information requested is for AI / ML, or the UE may use a well-known AI / ML DNN to request LADN information for an LADN PDU session which may be used for AI / ML. For instance, the UE may request this information using NAS messages or RRC messages.

[0096] In certain examples, the UE may receive LADN information for AI / ML, which may include a list of TAIs indicating where the LADN PDU session can be obtained for AI / ML. Once in this TAI, the UE may establish a PDU session for LADN for AI / ML and then obtain AI / ML data.

[0097] Further to the above, note that LADN PDU session may be considered to act as a local PDU session which is closer to the UE’s location, and hence may improve the overall latency in sharing the data to the UE.

[0098] In certain examples, the network may provide this information (i.e., LADN information) to the UE optionally if the UE indicates support for LADN AI / ML connection or for AI / ML, or based on subscription information. In some examples, the LADN information for AI / ML may be provided to the UE via NAS messages or RRC messages. The network may provide this information to the UE optionally if the subscription information indicates that this is permissible (or required) for the UE in question, optionally where the subscription information may also contain a list of TAIs (or general location information) that may define where the UE need be in order to provide this information to the UE.

[0099] Figure 5 is a block diagram illustrating an exemplary network entity 200 (or electronic device, or network node etc.) that may be used in examples of the present disclosure. For example, a UE, device, network entity, network node, network function, network etc. as described in any of the embodiments / examples disclosed above may be implemented by or comprise network entity 200 (or be in combination with network entity 200). For example, AI / ML NF 100 may be implemented by or in combination with, or comprise, network entity 200.

[00100] The network entity 200 comprises a controller 205 (or at least one processor) and at least one of a transmitter 201, a receiver 203, or a transceiver (not shown).

[00101] For example: controller 205 may be arranged to control the network entity 200 to perform any of the one or more features, operations or functions disclosed in relation to a network entity above; transmitter 201 may be arranged to transmit any one or more of the information, signals, data etc. mentioned above; and receiver 203 may be arranged to receive any one or more of the information, signals, data etc. mentioned above. The person skilled in the art would understand how such a network entity 200 in accordance with anyone or more example / embodiment disclosed herein may be provided.

[00102] For all of the examples / aspects / embodiments etc. described above / herein, it should be considered that the corresponding features / operations apply in any order or combination, and that furthermore there exists the possibility to omit one or more features / operations. Also, references herein to the “network” may refer to at least one of the RAN, AMF, SMF, UDM, NEF, PCF, the new defined entity (non-limitingly termed AI / ML NF 100), or any other entity in the 5GC and is not limited to particular nodes / entities. Of course, it will also be appreciated that the network may be defined differently. Furthermore, network signalling to provide any of the abovedescribed information may, as non-limiting examples, include RRC or NAS messages or system information. It will be appreciated that, optionally, network entities may first share necessary information amongst them / each other and then a recipient entity may share the information with the UE using the appropriate signalling.

[00103] Moreover, for all of the proposals above, the proposals apply to at least LTE, NR, NR NTN or loT NTN (note this list is merely to give some examples and should not be seen as limiting), including any related signalling / messages on any of the inferences X2, Xn, NG, S1, F1, etc (again, this list is merely to give some examples and should not be seen as limiting).It will be appreciated that, in each example / embodiment / aspect etc. described above, one or more features or operations may be omitted, modified or moved (e.g., to change the order of the features or the operations), if desired and appropriate.

[00104] Additionally, regarding al of the above, one or more features or operations etc. from any example / embodiment may be combined with features or operations from any other example / embodiment. That is, the present disclosure should be considered to include all combinations of examples / embodiments disclosed herein, as appropriate, as well as combinations of individual features within and between each example / embodiment, as appropriate.

[00105] The techniques described herein may be implemented using any suitably configured apparatus and / or system. Such an apparatus and / or system may be configured to perform a method according to any aspect, embodiment or example disclosed herein. Such an apparatus may comprise one or more elements, for example one or more of receivers, transmitters, transceivers, processors, controllers, modules, units, and the like, each element configured to perform one or more corresponding processes, operations and / or method steps for implementing the techniques described herein. For example, an operation / function of X may be performed by a module configured to perform X (or an X-module). The one or more elements may be implemented in the form of hardware, software, or any combination of hardware and software.

[00106] It will be appreciated that examples of the present disclosure may be implemented in the form of hardware, software or any combination of hardware and software. Any such software may be stored in the form of volatile or non-volatile storage, for example a storage device like a ROM, whether erasable or rewritable or not, or in the form of memory such as, for example, RAM, memory chips, device or integrated circuits or on an optically or magnetically readable medium such as, for example, a CD, DVD, magnetic disk or magnetic tape or the like.

[00107] It will be appreciated that the storage devices and storage media are embodiments of machine-readable storage that are suitable for storing a program or programs comprising instructions that, when executed, implement certain examples of the present disclosure. Accordingly, certain examples provide a program comprising code for implementing a method, apparatus or system according to any example, embodiment and / or aspect disclosed herein, and / or a machine-readable storage storing such a program. Still further, such programs may be conveyed electronically via any medium, for example a communication signal carried over a wired or wireless connection.

[00108] While the invention has been shown and described with reference to certain examples, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the scope of the invention.

[00109] The reader's attention is directed to all papers and documents which are filed concurrently with or previous to this specification in connection with this application and which are open to public inspection with this specification, and the contents of all such papers and documents are incorporated herein by reference.

[00110] Annex The following definition of terms is found in TS 23.247 [2]: 5GC Individual MBS traffic delivery: 5G CN receives a single copy of MBS data packets and delivers separate copies of those MBS data packets to individual UEs via per-UE PDU sessions, hence for each such UE one PDU session is required to be associated with a Multicast MBS Session. 5GC Shared MBS traffic delivery: 5G CN receives a single copy of MBS data packets and delivers a single copy of those MBS data packets to a RAN node. Area Session Identifier: A unique identifier within an MBS Session used for an MBS session with location dependent content. When present, the Area Session ID, together with the TMGI, is used to uniquely identify the data flow of an MBS Session in a specific MBS service area. Associated PDU Session: A PDU Session associated to a multicast MBS session that is used for 5GC Individual MBS traffic delivery method and for signalling related to a user's participation in a multicast MBS session such as join and leave requests. Associated QoS Flow: A unicast QoS Flow that belongs to the associated PDU Session and is used for 5GC Individual MBS traffic delivery method. The associated QoS Flow is mapped from a multicast QoS Flow in a multicast MBS session. Broadcast communication service: A 5GS communication service in which the same service and the same specific content data are provided simultaneously to all UEs in a geographical area (i.e. all UEs in the broadcast coverage area are authorized to receive the data). NOTE 1: For the broadcast communication service, the content provider and network may not be aware whether the authorized UEs are actually receiving the data being delivered. Broadcast MBS session: An MBS session to deliver the broadcast communication service. A broadcast MBS session is characterised by the content to send and the geographical area where to distribute it. Broadcast service area: The area within which data of one or multiple Broadcast MBS session(s) are sent. MBS QoS Flow: The finest granularity for QoS forwarding treatment for MBS data. Providing different QoS forwarding treatment requires separate MBS QoS Flows in 5GS supporting MBS. MBS Service Announcement: Mechanism to allow users to be informed about the available MBS services. MBS session: A multicast MBS session or a broadcast MBS session. MBS service area: The area within which data of one Multicast or Broadcast MBS session may 5 be sent. For location dependent MBS, for each MBS service area, an Area Session ID, which is unique per MBS Session ID, is allocated and the same location dependent content data for an MBS session is delivered to the UE(s) within an MBS service area. Multicast communication service: A 5GS communication service in which the same service and 10 the same specific content data are provided simultaneously to a dedicated set of UEs (i.e. not all UEs in the coverage of the MBS service area are authorized to receive the data). NOTE 2: For multicast communication service, the content provider and network can be aware whether the authorized UEs are actually receiving the data being delivered. 15 Multicast MBS session: An MBS session to deliver the multicast communication service. A multicast MBS session is characterised by the content to send, by the list of UEs that may receive the service and optionally by a geographical area where to distribute it. Acronyms and Definitions 3GPP 5G 5GC 3rd Generation Partnership Project 5th Generation 5G Core 5 5QI 5G QoS Identifier 5GS 5G System 5GSM 5G System Session Management 5GMM 5G System Mobility Management AF Application Function 10 Al Artificial Intelligence AM Acknowledged Mode AMF Access and Mobility Management Function AS Application Server ASP Application Service Provider 15 AUSF Authentication Server Function CDN Content Delivery Network DCAF Data Collection Application Function DNAI Data Network Access Identifier DNN Data Network Name 20 DNS Domain Name Server DRB Data Radio Bearer eNB Evolved Node B FEC Forward Error Correction FQDN Fully Qualified Domain Name 25 GBR Guaranteed Bit Rate gNB Next generation Node B GPSI Generic Public Subscription Identifier IAB Integrated Access and Backhaul ID Identity / ldentifier 30 HoT Industrial Internet of Things IMEI International Mobile Equipment Identities IP Internet Protocol l-SMF Intermediate SMF LADN Local Area Data Network 35 LL SSM Lower Layer SSM MBMS Multimedia Broadcast / Multicast Service MBS Multicast / Broadcast Service MBSF Multicast / Broadcast Service Function MBSTF Multicast / Broadcast Service Transport Function 40 MB-SMF Multicast / Broadcast Session Management Function MB-UPF Multicast / Broadcast User Plane Function ML Machine Learning MME Mobility Management Entity MN Master Node 45 MNO Mobile Network Operator MT Mobile Termination NAS Non-Access Stratum NEF Network Exposure Function NRF Network Repository Function 50 NG-RAN Next Generation Radio Access Network NG-eNB Next Generation eNB NSA Non-Standalone NSSF Network Slice Selection Function NTN Non-Terrestrial Networks 55 NW Network NWDAF Network Data Analytics Function OS Operating System OSAPP OS Application PCF Policy Control Function 60 PCO Protocol Configuration Options PDR Packet Detection Rule PDU Protocol Data Unit PTM Point To Multipoint PTP Point to Point QFI QoS Flow Identifier (ID) 5 QoS Quality of Service RACH Random Access Channel RAN Radio Access Network RSD Route Selection Descriptor SA Standalone 10 SDAP Service Data Adaptation Protocol SDU Service Data Unit SIM Subscriber Identity Module SLA Service Level Agreement SM Session Management 15 SMF Session Management Function SN Secondary Node S-NSSAI Single Network Slice Selection Assistance Information SSB Synchronization Signal Block SSM Source Specific IP Multicast address 20 SSC Session and Service Continuity SUPI Subscription Permanent Identifier TAI Tracking Area Identity TE Terminal Equipment TM Transparent Mode 25 TMGI Temporary Mobile Group Identity TS Technical Specification UDM Unified Data Manager UDR Unified Data Repository UE User Equipment 30 UL Uplink UM Unacknowledged Mode UP User Plane UPF User Plane Function URLLC Ultra-Reliable and Low-Latency Communication 35 URSP UE Route Selection Policy 24 07 25

Claims

1. A first entity included in a communications network, the first entity comprising: a transmitter;5 a receiver; anda controller configured to:receive, from a second entity included in the communications network, first information or signalling associated with transfer of artificial intelligence / machine learning (AI / ML) data; and10 transmit first AI / ML data to the second entity using multicast-broadcast services(MBS) ora local area data network (LADN), based on the first information or signalling;wherein the first information or signalling comprises an indication that the second entity supports MBS or an indication that the second entity supports LADN AI / ML connection.15 2. The first entity of claim 1, wherein the first information or signalling comprises the indicationthat the second entity supports MBS; andwherein the controller is configured to:based on receiving the indication, transmit, to the second entity, second information or signalling for configuring the second entity to request the first AI / ML data; and20 receive, from the second entity, a request for the first AI / ML data; andwherein the first AI / ML data is transmitted using MBS in response to the request.

3. The first entity of claim 1, wherein the first information or signalling comprises a request for the first AI / ML data; and25 wherein the first AI / ML data is transmitted to the second entity in response to the request.

4. The first entity of claim 2 or claim 3, wherein the request comprises information identifying one or more AI / ML models and / or information associated with one or more AI / ML models; andwherein the first AI / ML data comprises the identified one or more AI / ML models and / or 30 information associated with one or more AI / ML models.

5. The first entity of any previous claim, wherein the controller is configured to: transmit, to the second entity, one or more parameters for receiving the first AI / ML data using MBS; and / or35 transmit, to the second entity, an indication of presence of MBS data for a AI / ML service; andwherein the first AI / ML data is transmitted to the second entity using MBS.

6. The first entity of claim 5, wherein the one or more parameters are transmitted using dedicated signalling or system information, or signalling using the application layer; and / or40 wherein the one or more parameters comprises one or more of:at least one Temporary Mobile Group Identity (TMGI),24 07 25at least one Internet Protocol (IP) address, or information on an area of multicast.

7. The first entity of any previous claim, wherein the first entity is a network function (NF)5 included in a 5G core (5GC) connected to a radio access network (RAN), and is connected to one or more MBS-related function; andwherein the controller is configured to:collect data for AI / ML related to the RAN, the collected data including the first AI / ML data; and10 provide the first AI / ML data into a MBS framework for transmission to the second entity.

8. The first entity of claim 7, wherein the controller is configured to:collect data regarding delivery modes of previous MBS traffic and / or status of the RAN; and provide, to the connected one or more MBS-related function, a recommendation on a delivery15 mode decision for the second entity and / or a recommendation to switch a delivery mode.

9. The first entity of claim 7 or claim 8, wherein a protocol data unit (PDU) session is established between the second entity and the first entity for transfer of the AI / ML data.20 10. The first entity of claim 1, wherein the controller is configured to:deploy the LADN in an area for transferring the AI / ML data to the second entity; and transmit the first AI / ML data to the second entity using the LADN when a protocol data unit (PDU) session is established with the second entity.25 11. The first entity of claim 1, wherein the first information or signalling comprises the indicationthat the second entity supports LADN AI / ML connection and a request for LADN information; and wherein the controller is configured to:in response to receiving the first information or signalling, transmit the LADN information to the second entity, wherein the LADN information is for use in establishing the PDU session or the30 LADN information is requested for transfer of the first AI / ML data.

12. The first entity of claim 11, wherein the LADN information comprises a list of tracking area identifiers (TAIs) indicating where the PDU session can be obtained.35 13. A second entity included in a communications network, the second entity comprising:a transmitter;a receiver; anda controller configured to:transmit, to a first entity included in the communications network, first information or40 signalling associated with transfer of artificial intelligence / machine learning (AI / ML) data; and24 07 25receive first AI / ML data to the second entity using multicast-broadcast services (MBS) or a local area data network (LADN), based on the first information or signalling;wherein the first information or signalling comprises an indication that the second entity supports MBS or an indication that the second entity supports LADN AI / ML connection.

514. The second entity of claim 13, wherein the first information or signalling comprises the indication that the second entity supports MBS; andwherein the controller is configured to:receive, from the first entity, second information or signalling for configuring the second entity10 to request the first AI / ML data; andtransmit, to the first entity, a request for the first AI / ML data based on the second information or signalling; andwherein the first AI / ML data is received using MBS.15 15. The second entity of claim 13, wherein the first information or signalling comprises a requestfor the first AI / ML data; andwherein the first AI / ML data is received using MBS.

16. The second entity of claim 14 or claim 15, wherein the request comprises information 20 identifying one or more AI / ML models and / or information associated with one or more AI / ML models;andwherein the first AI / ML data comprises the identified one or more AI / ML models and / or information associated with one or more AI / ML models.25 17. The second entity of any one of claims 13 to 16, wherein the controller is configured to:receive, from the first entity, one or more parameters for receiving the first AI / ML data using MBS; and / orreceive, from the first entity, an indication of presence of MBS data for a AI / ML service; and wherein the first AI / ML data is received from the first entity using MBS based on the one or 30 more parameters and / or the indication of presence.

18. The second entity of claim 17, wherein the one or more parameters are transmitted using dedicated signalling or system information, or signalling using the application layer; and / orwherein the one or more parameters comprises one or more of:35 at least one Temporary Mobile Group Identity (TMGI),at least one Internet Protocol (IP) address, or information on an area of multicast.

19. The second entity of claim 17 or claim 18, wherein the controller is configured to join an MBS 40 session for receiving the first AI / ML data based on one of:24 07 25determining the second entity is in an area where MBS is available for transfer of AI / ML data, based on the one or more parameters and / or the indication of presence;determining a current location of the second entity is within an area indicated in the indication of presence;receiving the indication of presence in response to a request, transmitted from the second entity to the first entity, for the first AI / ML data; ordetermining a current location of the second entity is within a satellite coverage area where the MBS is available, based on satellite ephemeris information, the one or more parameters and / or the indication of presence.

20. The second entity of claim 13, wherein the controller is configured to:establish a protocol data unit (PDU) session with the first entity for receiving the first AI / ML data using MBS, based on one or more MBS parameters received from the first entity or from a third entity in the communications network.

21. The second entity of claim 13, wherein the first information or signalling comprises the indication that the second entity supports LADN AI / ML connection and a request for LADN information; andwherein the controller is configured to:receive the LADN information from the first entity, wherein the LADN information is for use in establishing a LADN protocol data unit (PDU) session with the first entity or the LADN information is requested for transfer of the first AI / ML data.

22. The second entity of claim 21, wherein the LADN information comprises a list of tracking area identifiers (TAIs) indicating where the LADN PDU session can be obtained;wherein the controller is configured to establish the LAD PDU session based on determining a current location of the second entity is within a TAI included in the list of TAIs; andwherein the first AI / ML data is received via the LADN PDU session.

23. The first entity of any of claims 1 to 6 or the second entity of any of claims 13 to 22, wherein: the first entity is one of 5G core (5GC), radio access network (RAN) or a network function (NF); and / orthe second entity is a user equipment (UE) or a plurality of UEs.

24. A method of a first entity included in a communications network, the method comprising: receiving, from a second entity included in the communications network, first information or signalling associated with transfer of artificial intelligence / machine learning (AI / ML) data; andtransmitting first AI / ML data to the second entity using multicast-broadcast services (MBS) or a local area data network (LADN), based on the first information or signalling;wherein the first information or signalling comprises an indication that the second entity supports MBS or an indication that the second entity supports LADN AI / ML connection.

25. A method of a second entity included in a communications network, the method comprising: transmitting, to a first entity included in the communications network, first information or signalling associated with transfer of artificial intelligence / machine learning (AI / ML) data; and5 receiving first AI / ML data to the second entity using multicast-broadcast services (MBS) or alocal area data network (LADN), based on the first information or signalling;wherein the first information or signalling comprises an indication that the second entity supports MBS or an indication that the second entity supports LADN AI / ML connection.24 07 25

Images