NW initiated PDU session management for ML model data transfer

Abstract

A method includes identifying, by a first apparatus hosting a machine learning (ML) Model Data, that a User Equipment (UE) in a radio resource control (RRC) connected state is a candidate for receiving the ML Model data. The first apparatus transmits, based on the identifying, a first trigger message to a second apparatus to setup at least one of a dedicated protocol data unit (PDU) session or a dedicated quality of service (QoS) flow for the UE, receives from the second apparatus, a PDU session setup request or a QoS flow setup request. The first apparatus sets up a data radio bearer associated with the PDU session or QoS flow, and transmits the ML Model Data over the data radio bearer based upon a determination to deliver the ML Model Data to the UE.

Description

NW INITIATED PDU SESSION MANAGEMENT FOR ML MODEL DATA TRANSFERFIELD

[0001] Various example embodiments relate generally to wireless networks and, more particularly, for a method and apparatus for network (NW) initiated Protocol Data Unit (PDU) session management for machine learning (ML) Model Data transfer.BACKGROUND

[0002] To enable data transfer between a user equipment (UE) and Application Server (AS), resources such as Data Radio Bearer (DRB) over the air interface between the UE and Radio Access Network (RAN), NG-U tunnels between the RAN and Core Network (CN) and any tunnels between nodes of split-architecture RAN have to be setup.

[0003] Conventionally, the setup of these resources are triggered by the Core Network (e.g., a session management function (SMF)) or the UE. These are procedures for UE-Initiated PDU Session Management or Network-Initiated PDU Session Management. Data to be transferred is in the UE or in the application server. Once the resources are setup, data is exchanged between UE and Application Server.SUMMARY

[0004] In an aspect of the present disclosure, a method includes identifying, by a first apparatus hosting a machine learning (ML) Model Data, that a User Equipment (UE) in a radio resource control (RRC) connected state is a candidate for receiving the ML Model data. The first apparatus transmits, based on the identifying, a first trigger message to a second apparatus to setup at least one of a dedicated protocol data unit (PDU) session or a dedicated quality of service (QoS) flow for the UE, receives from the second apparatus, a PDU session setup request or a QoS flow setup request. The first apparatus sets up a data radio bearer associated with the PDU session or QoS flow, and transmits the ML Model Data over the data radio bearer based upon a determination to deliver the ML Model Data to the UE.

[0005] In an aspect of the method, the first trigger message includes an indication that the requested PDU session or QoS flow is for delivery of an ML Model data.

[0006] In an aspect of the method, the first trigger message includes an ML Model Data ID.

[0007] In an aspect of the method, the method further includes including, by the first apparatus, in the first trigger message to the second apparatus a binding identifier (binding ID); receiving, by the first apparatus, from the second apparatus a PDU session setup request of QoS flow setup request including the binding identifier; determining, by the first apparatus, based on the binding identifier, a mapping between the received PDU session setup request or QoS flow setup request and the first trigger message; and deciding based on the determining, by the first apparatus, to transmits the identified UE the ML Model Data over a data radio bearer associated with the PDU session or QoS flow.

[0008] In an aspect of the method, the binding ID is the ML Model Data ID.

[0009] In an aspect of the method, the method further includes determining, by the first apparatus, that a PDU session setup request or QoS flow setup request received from the second apparatus is for delivery of an ML Model Data; and setting up, by the first apparatus, a data radio bearer corresponding to the requested PDU session or QoS flow without setting up an associated tunnel between the first apparatus and a third apparatus.

[0010] In an aspect of the method, the determining that a PDU session setup request or QoS flow setup request received from the second apparatus is for delivery of an ML Model Data includes receiving in the PDU session setup request or the QoS flow setup request a binding ID or an ML Model Data ID.

[0011] In an aspect of the method, the transmission to the second apparatus in response to the received PDU session setup request or QoS flow setup request includes at least one of: an indicator to not setup the tunnel between the first apparatus and the third apparatus, or not including a downlink tunnel endpoint for a tunnel between the first apparatus and the third apparatus.

[0012] In an aspect of the method, the method further includes detecting, by the first apparatus, the end of delivery of the ML Model Data to the UE; transmitting, to the second apparatus, a notification indicating the end of delivery or a second trigger message to release the PDU session or to release the QoS flow; receiving, by the first apparatus, a PDU session release request of QoS flow release request from the second apparatus; and releasing the data radio bearer associated with the said PDU session or QoS flow.

[0013] In an aspect of the method, the notification or the second trigger message includes at least one of the ML Model Data ID or the binding identifier.

[0014] In an aspect of the method, the PDU session release request or the QoS flow release request includes at least one of the ML Model Data ID or the binding identifier.

[0015] In an aspect of the method, the method further includes determining, by the first apparatus, an ML Model data size or an ML Model data type associated with the ML Model data; and identifying, by the first apparatus, the UE based on the determined ML Model data size or ML Model data type.

[0016] In an aspect of the method, the first apparatus is an NG-RAN node of a 5G cellular network.

[0017] In an aspect of the method, the first apparatus is a gNB-CU CP of a gNB split between a gNB-CU CP and one or more gNB-CU UP(s).

[0018] In an aspect of the method, the method further includes receiving, from a gNB-CU UP, a notification of availability of ML Model data in the gNB-CU UP; and identifying a UE and sending the first trigger message based on receiving this notification.

[0019] In an aspect of the method, the method further includes receiving from the gNB-CU UP at least one of an ML Model Data ID or an ML Model Data Size or an ML Model Data type.

[0020] In an aspect of the method, the method includes sending to the gNB-CU UP a request to start the ML Model data delivery over an Fl-U tunnel associated with a context of the identified UE.

[0021] In an aspect of the method, the method further includes receiving from the gNB-CU UP an indication of end of ML Model Data delivery; and sending the notification of end delivery to the second apparatus or sending the second trigger message to the second apparatus based on the receiving of this indication from gNB-CU UP.

[0022] In an aspect of the method, the second apparatus is a core network entity and the core network entity is an Access Management Function (AMF) or a Session Management Function (SMF) of a 5G cellular network.

[0023] In an aspect of the method, the third apparatus is another core network entity and wherein the another core network entity is a user plane function (UPF) of a 5G cellular network.

[0024] In an aspect of the method, the first trigger message, the second trigger message and the notification message are NG Application Protocol (NGAP) messages.

[0025] In an aspect of the present disclosure, a method includes receiving, by a first core network entity, from a network access node, a first trigger message to setup at least one of adedicated protocol data unit (PDU) session or a dedicated quality of service (QoS) flow; and transmitting, by the first core network entity, a PDU session setup request or a QoS flow setup request to the network access node.

[0026] In an aspect of the method, the first trigger message includes an indication that the requested PDU session or QoS flow is for delivery of a machine learning (ML) Model data.

[0027] In an aspect of the method, the first trigger message includes an ML Model Data ID.

[0028] In an aspect of the method, the method further includes receiving, by the first core network entity, in the first trigger message from the network access node a binding identifier (binding ID); and transmitting, by the first core network entity, to the network access node, in response to the first trigger message, a PDU session setup request of QoS flow setup request including the binding ID.

[0029] In an aspect of the method, the binding ID is the ML Model Data ID.

[0030] In an aspect of the method, the method further includes determining, by the first core network entity, from receiving the first trigger message, that the requested PDU session setup request or QoS flow setup request is for delivery of an ML Model Data; and transmitting, by the first core network entity, the PDU Session setup request or QoS flow setup request without an Uplink tunnel endpoint of a second core network entity.

[0031] In an aspect of the method, the method further includes including, in the PDU session setup request or the QoS flow setup request a Non Access Stratum (NAS) container towards the UE indicating that the PDU session setup or QoS flow setup is associated with an ML Model data delivery.

[0032] In an aspect of the method, the method further includes including, in the PDU session setup request or the QoS flow setup request a Non Access Stratum (NAS) container towards the UE indicating an ML Model Data ID.

[0033] In an aspect of the method, the method further includes receiving, by the first core network entity from the network access node, a notification indicating the end of the ML Model Data delivery or a second trigger message to release the PDU session or release the QoS flow; and transmitting, by the first core network entity, to the network access node in response to the notification or second trigger message, a PDU session release request or QoS flow release request.

[0034] In an aspect of the method, the notification or the second trigger message includes at least one of the ML Model Data ID or the binding identifier.

[0035] In an aspect of the method, the PDU session release request or the QoS flow release request includes at least one of the ML Model Data ID or the binding identifier.

[0036] In an aspect of the method, the network access node is an NG-RAN node of a 5G cellular network.

[0037] In an aspect of the method, the first core network entity is an Access Management Function (AMF) or a Session Management Function (SMF) of a 5G cellular network.

[0038] In an aspect of the method, the second core network entity is a user plane function (UPF) of a 5G cellular network.

[0039] In an aspect of the method, the first trigger message, the second trigger message and the notification message are NG Application Protocol (NGAP) messages.

[0040] In an aspect of the present disclosure, a method includes receiving, by a user equipment (UE) from a core network entity, a protocol data unit (PDU) Session setup request or quality of service (QoS) flow setup request comprising an indication that the PDU session setup request or the QoS flow setup request is for delivery of a machine learning (ML) Model Data; and storing, by the UE, data received over a radio bearer associated with the PDU session or the QoS flow as data of an ML model.

[0041] In an aspect of the method, the indication is an ML Model Data ID.

[0042] In an aspect of the method, the method further includes associating and storing, by the UE, the ML Model Data ID with the data received over the radio bearer associated with the PDU session or the QoS flow.

[0043] In an aspect of the method, the receiving a PDU session setup request or QoS flow setup request comprises receiving a PDU Session setup request or QoS flow setup request.

[0044] In an aspect of the method, the receiving an indication that the PDU session setup request or the QoS flow setup request is for delivery of an ML Model Data includes receiving the indication in a Non Access Stratum (NAS) message from a core network entity.

[0045] In an aspect of the method, the receiving an indication that the PDU session setup request or the QoS flow setup request is for delivery of an ML Model Data, comprises receiving the indication in a Radio Resource Control (RRC) message from a network access node.

[0046] In an aspect of the method, the network access node is an NG-RAN node of a 5G cellular network.

[0047] In an aspect of the method, the core network entity is an Access Management Function (AMF) or a Session Management Function (SMF) of a 5G cellular network.

[0048] In an aspect of the present disclosure, a UE includes at least one processor and at least one memory storing instructions which, when executed by the at least one processor, causes the UE at least to perform any of the foregoing methods.

[0049] In an aspect of the present disclosure, an apparatus includes at least one processor and at least one memory storing instructions which, when executed by the at least one processor, causes the apparatus at least to perform any of the foregoing methods.

[0050] In an aspect of the present disclosure, a processor-readable medium storing instructions which, when executed by at least one processor of an apparatus, cause the apparatus at least to perform any of the foregoing methods.

[0051] According to some aspects, there is provided the subject matter of the independent claims. Some further aspects are defined in the dependent claims.BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Some example embodiments will now be described with reference to the accompanying drawings.

[0053] FIG. 1 is a diagram of an example embodiment of wireless networking between a network system and a user equipment (UE), according to one illustrated aspect of the disclosure;

[0054] FIG. 2 is a diagram of example components of a network system, according to one illustrated aspect of the disclosure;

[0055] FIG. 3 is a diagram of an example interaction between a UE, a RAN and a Core Network, according to one illustrated aspect of the disclosure;

[0056] FIG. 4 is a diagram of an example embodiment of signals and operations for a QoS flow setup among a UE, RAN and Core Network, according to one illustrated aspect of the disclosure;

[0057] FIG. 5 is a diagram of an example embodiment of signals and operations for a QoS flow release among a UE, RAN and Core Network, according to one illustrated aspect of the disclosure;

[0058] FIG. 6 is a diagram of an example embodiment of signals and operations for a RAN initiated ML Model Data associated resource setup trigger among a RAN and Core Network, according to one illustrated aspect of the disclosure;

[0059] FIG. 7 is a diagram of an example embodiment of signals and operations for a RAN initiated ML Model Data associated resource setup trigger among a RAN and Core Network, according to another illustrated aspect of the disclosure;

[0060] FIG. 8 is a diagram of an example embodiment of signals and operations for a RAN initiated ML Model Data associated resource setup trigger among a RAN and Core Network, according to another illustrated aspect of the disclosure;

[0061] FIG. 9 is a diagram of an example embodiment of signals and operations for a RAN initiated ML Model Data associated resource setup trigger among a RAN and Core Network, according to another illustrated aspect of the disclosure;

[0062] FIG. 10 is a diagram of an example embodiment of signals and operations for a Core initiated ML Model Data flow associated QoS flow setup among a RAN and Core Network, according to another illustrated aspect of the disclosure;

[0063] FIG. 11 is a diagram of an example embodiment of signals and operations for end to end ML Model Data transfer among a UE, RAN and Core Network, according to one illustrated aspect of the disclosure; and

[0064] FIG. 12 is a diagram of an example block diagram of a wireless station or node (e.g., network node (such as gNB), user node or UE, relay node, or other node), according to one illustrated aspect of the present disclosure.DETAILED DESCRIPTION

[0065] In the following description, certain specific details are set forth in order to provide a thorough understanding of disclosed aspects. However, one skilled in the relevant art will recognize that aspects may be practiced without one or more of these specific details or with other methods, components, materials, etc. In other instances, well-known structures associated with transmitters, receivers, or transceivers have not been shown or described in detail to avoid unnecessarily obscuring descriptions of the aspects.

[0066] Reference throughout this specification to “one aspect” or “an aspect” means that a particular feature, structure, or characteristic described in connection with the aspect is includedin at least one aspect. Thus, the appearances of the phrases “in one aspect” or “in an aspect” in various places throughout this specification are not necessarily all referring to the same aspect. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more aspects.

[0067] Embodiments described in the present disclosure may be implemented in wireless networking apparatuses, such as, without limitation, apparatuses utilizing Worldwide Interoperability for Microwave Access (WiMAX), Global System for Mobile communications (GSM, 2G), GSM EDGE radio access Network (GERAN), General Packet Radio Service (GRPS), Universal Mobile Telecommunication System (UMTS, 3G) based on basic wideband-code division multiple access (W-CDMA), high-speed packet access (HSPA), Long Term Evolution (LTE), LTE- Advanced, enhanced LTE (eLTE), 5G New Radio (5G NR), 5G Advance, 6G (and beyond) and 802.1 lax (Wi-Fi 6), among other wireless networking systems. The term ‘eLTE’ here denotes the LTE evolution that connects to a 5G core. LTE is also known as evolved UMTS terrestrial radio access (EUTRA) or as evolved UMTS terrestrial radio access network (EUTRAN).

[0068] The present disclosure may use the term “serving network device” to refer to a network node or network device (or a portion thereof) that services a UE. As used herein, the terms “transmit to,” “receive from,” and “cooperate with,” (and their variations) include communications that may or may not involve communications through one or more intermediate devices or nodes. The term “acquire” (and its variations) includes acquiring in the first instance or reacquiring after the first instance. The term “connection” may mean a physical connection or a logical connection.

[0069] The present disclosure uses 5G NR as an example of a wireless network and may use smartphones and / or extended reality headsets as an example of UEs. It is intended and shall be understood that such examples are merely illustrative, and the present disclosure is applicable to other wireless networks and user equipment.

[0070] FIG. 1 is a diagram depicting an example of wireless networking between a network system 100 and a user equipment (UE) 150. The network system 100 may include one or more network nodes 120, one or more servers 110, and / or one or more network equipment 130 (e.g., test equipment). The network nodes 120 will be described in more detail below. As used herein, the term “network apparatus” may refer to any component of the network system 100, such as the server 110, the network node 120, the network equipment 130, any component(s) of the foregoing, and / or any other component(s) of the network system 100. Examples of network apparatusesinclude, without limitation, apparatuses implementing aspects of 5G NR, among others. The present disclosure describes embodiments related to 5GNR and embodiments that involve aspects defined by 3rd Generation Partnership Project (3GPP). However, it is contemplated that embodiments relating to other wireless networking technologies are encompassed within the scope of the present disclosure.

[0071] The following description provides further details of examples of network nodes. In a 5G NR network, a gNodeB (also known as gNB) may include, e.g., a node that provides new radio (NR) user plane and control plane protocol terminations towards the UE and that is connected via a NG interface to the 5G core (5GC), e.g., according to 3GPP TS 38.300 V16.6.0 (2021-06) section 3.2, which is hereby incorporated by reference herein.

[0072] A gNB supports various protocol layers, e.g., Layer 1 (LI) - physical layer, Layer 2 (L2), and Layer 3 (L3).

[0073] The layer 2 (L2) of NR is split into the following sublayers: Medium Access Control (MAC), Radio Link Control (RLC), Packet Data Convergence Protocol (PDCP) and Service Data Adaptation Protocol (SDAP), where, e.g.: o The physical layer offers to the MAC sublayer transport channels; o The MAC sublayer offers to the RLC sublayer logical channels; o The RLC sublayer offers to the PDCP sublayer RLC channels; o The PDCP sublayer offers to the SDAP sublayer radio bearers; o The SDAP sublayer offers to 5GC quality of service (QoS) flows; o Control channels include broadcast control channel (BCCH) and physical control channel (PCCH).

[0074] Layer 3 (L3) includes, e.g., radio resource control (RRC), e.g., according to 3GPP TS 38.300 V16.6.0 (2021-06) section 6, which is hereby incorporated by reference herein.

[0075] A gNB central unit (gNB-CU) includes, e.g., a logical node hosting, e.g., radio resource control (RRC), service data adaptation protocol (SDAP), and packet data convergence protocol (PDCP) protocols of the gNB or RRC and PDCP protocols of the en-gNB, that controls the operation of one or more gNB distributed units (gNB-DUs). The gNB-CU terminates the Fl interface connected with the gNB-DU. A gNB-CU may also be referred to herein as a CU, a central unit, a centralized unit, or a control unit.

[0076] A gNB Distributed Unit (gNB-DU) includes, e.g., a logical node hosting, e.g., radio link control (RLC), media access control (MAC), and physical (PHY) layers of the gNB or en- gNB, and its operation is partly controlled by the gNB-CU. One gNB-DU supports one or multiple cells. One cell is supported by only one gNB-DU. The gNB-DU terminates the Fl interface connected with the gNB-CU. A gNB-DU may also be referred to herein as DU or a distributed unit.

[0077] As used herein, the term “network node” may refer to any of a gNB, a gNB-CU, or a gNB-DU, or any combination of them. A RAN (radio access network) node or network node such as, e.g., a gNB, gNB-CU, or gNB-DU, or parts thereof, may be implemented using, e.g., an apparatus with at least one processor and / or at least one memory with processor-readable instructions (“program”) configured to support and / or provision and / or process CU and / or DU related functionality and / or features, and / or at least one protocol (sub-)layer of a RAN (radio access network), e.g., layer 2 and / or layer 3. Different functional splits between the central and distributed unit are possible. An example of such an apparatus and components will be described in connection with FIG. 5 below.

[0078] The gNB-CU and gNB-DU parts may, e.g., be co-located or physically separated. The gNB-DU may even be split further, e.g., into two parts, e.g., one including processing equipment and one including an antenna. A central unit (CU) may also be called baseband unit / radio equipment controller / cloud-RAN / virtual-RAN (BBU / REC / C-RAN / V-RAN), open-RAN (O- RAN), or part thereof. A distributed unit (DU) may also be called remote radio head / remote radio unit / radio equipment / radio unit (RRH / RRU / RE / RU), or part thereof. Hereinafter, in various example embodiments of the present disclosure, a network node, which supports at least one of central unit functionality or a layer 3 protocol of a radio access network, may be, e.g., a gNB-CU. Similarly, a network node, which supports at least one of distributed unit functionality or a layer 2 protocol of the radio access network, may be, e.g., a gNB-DU.

[0079] A gNB-CU may support one or multiple gNB-DUs. A gNB-DU may support one or multiple cells and, thus, could support a serving cell for a user equipment (UE) or support a candidate cell for handover, dual connectivity, and / or carrier aggregation, among other procedures.

[0080] The user equipment (UE) 150 may be or include a wireless or mobile device, an apparatus with a radio interface to interact with a RAN (radio access network), a smartphone, an in-vehicle apparatus, an loT device, or a M2M device, among other types of user equipment. SuchUE 150 may include: at least one processor; and at least one memory including program code; where the at least one memory and the computer program code are configured to, with the at least one processor, cause the apparatus at least to perform certain operations, such as, e.g., RRC connection to the RAN. An example of components of a UE will be described in connection with FIG. 12. In embodiments, the UE 150 may be configured to generate a message (e.g., including a cell ID) to be transmitted via radio towards a RAN (e.g., to reach and communicate with a serving cell). In embodiments, the UE 150 may generate and transmit and receive RRC messages containing one or more RRC PDUs (packet data units). Persons skilled in the art will understand RRC protocol as well as other procedures a UE may perform.

[0081] With continuing reference to FIG. 1, in the example of a 5G NR network, the network system 100 provides one or more cells, which define a coverage area of the network system 100. As described above, the network system 100 may include a gNB of a 5G NR network or may include any other apparatus configured to control radio communication and manage radio resources within a cell. As used herein, the term “resource” may refer to radio resources, such as a resource block (RB), a physical resource block (PRB), a radio frame, a subframe, a time slot, a sub-band, a frequency region, a sub-carrier, a beam, etc. In embodiments, the network node 120 may be called a base station.

[0082] FIG. 1 provides an example and is merely illustrative of a network system 100 and a UE 150. Persons skilled in the art will understand that the network system 100 includes components not illustrated in FIG. 1 and will understand that other user equipment may be in communication with the network system 100.

[0083] FIG. 2 is a block diagram of example components of the network system 100 of FIG. 1. A 5G NR network may be described as an example of the network system 100, and it is intended that aspects of the following description shall be applicable to other types of network systems, as well. The network system may operate in accordance with the signals and connections shown in FIG. 1 such that the UE 150 is in communication with the network system 100 through the radio access network 225. Additionally, the network system may be divided into user plane components and functions and control plane components and functions, as shown and described herein. Unless indicated otherwise, the terms “component”, “function”, and “service” may be used interchangeably herein, and they may refer to and be implemented by instructions executed by one or more processors.

[0084] Example functions of the components are described below. The example functions are merely illustrative, and it shall be understood that additional operations and functions may be performed by the components described herein. Additionally, the connections between components may be virtual connections over service-based interfaces such that any component may communicate with any other component. In this manner, any component may act as a service “producer,” for any other component that is a service “consumer,” to provide services for network functions.

[0085] For example, a core network 210 is described in the control plane of the network system. The core network 210 may include an authentication server function (AUSF) 211, an access and mobility function (AMF) 212, and a session management function (SMF) 213. The core network 210 may also include a network slice selection function (NSSF) 214, a network exposure function (NEF) 215, a network repository function (NRF) 216, and a unified data management function (UDM) 217, which may include a uniform data repository (UDR) 224.

[0086] Additional components and functions of the core network 210 may include an application function 218, policy control function (PCF) 219, network data analytics function (NWDAF) 220, analytics data repository function (ADRF) 221, management data analytics function (MDAF) 222, and operations and management function (0AM) 223.

[0087] The user plane includes the UE 150, a radio access network (RAN) 225, a user plane function (UPF) 226, and a data network (DN) 227. The RAN 225 may include one or more components described in connection with FIG. 1, such as one or more network nodes. However, the RAN 225 may not be limited to such components. The UPF 226 provides connection for data being transmitted over the RAN 225. The DN 226 identifies services from service providers, Internet access, and third party services, for example.

[0088] The AMF 212 processes connection and mobility tasks. The AUSF 211 receives authentication requests from the AMF 212 and interacts with UDM 217 to authenticate and validate network responses for determination of successful authentication. The SMF 213 conducts packet data unit (PDU) session management, as well as manages session context with the UPF 226.

[0089] The NSSF 214 may select a network slicing instance (NSI) and determine the allowed network slice selection assistance information (NSSAI). This selection and determination is utilized to set the AMF 212 to provide service to the UE 150. The NEF 215 secures access tonetwork services for third parties to create specialized network services. The NRF 216 acts as a repository to store network functions to allow the functions to register with and discover each other.

[0090] The UDM 217 generates authentication vectors for use by the AUSF 211 and ADM 212 and provides user identification handling. The UDM 217 may be connected to the UDR 224 which stores data associated with authentication, applications, or the like. The AF 218 provides application services to a user (e.g., streaming services, etc.). The PCF 219 provides policy control functionality. For example, the PCF 219 may assist in network slicing and mobility management, as well as provide quality of service (QoS) and charging functionality.

[0091] The NWDAF 220 collects data (e.g., from the UE 150 and the network system) to perform network analytics and provide insight to functions that utilize the analytics in the providing of services. The ADRF 221 allows the storage, retrieval, and removal of data and analytics by consumers. The MDAF 222 provides additional data analytics services for network functions. The 0AM 223 provides provisioning and management processing functions to manage elements in or connected to the network (e.g., UE 150, network nodes, etc.).

[0092] FIG. 2 is merely an example of components of a network system, and variations are contemplated to be within the scope of the present disclosure. In embodiments, the network system may include other components not illustrated in FIG. 2. In embodiments, the network system may not include every component illustrated in FIG. 2. In embodiments, the components and connections may be implemented with different connections than those illustrated in FIG. 2. Such and other embodiments are contemplated to be within the scope of the present disclosure.

[0093] As mentioned above, to enable data transfer between a UE and Application Server (AS), resources such as Data Radio Bearer (DRB) over the air interface between the UE and RAN, NG-U tunnels between the RAN and core network and any tunnels between nodes of split-architecture RAN have to be setup. The term “RAN” and “RAN node” and “core network” and “core” may be used interchangeably throughout herein and above.

[0094] As networks continue to evolve with additional machine learning (ML) capability, there are new scenarios for which new paradigms / concepts in the network may be utilized. One such scenario is for ML Model Data Transfer which involves the transfer of ML Model Data between the node hosting the ML Model Data and the UE. When the ML Model data is available in the network, the network initiates the setup of the necessary resources by which the ML Modeldata can be transferred to the UE.

[0095] When the amount of ML Model Data is small enough, a control-plane transfer can be used. When the amount of ML Model Data is large, then a transfer mechanism using user-plane data transfer, still controlled by the control-plane procedures, may be utilized. In various embodiments, the network may trigger the resource setup as soon as ML Model Data is available. Alternatively, the ML Model Data can be transferred only when the UE explicitly requests it.

[0096] With ML Model Data Transfer, there are many possibilities as to which node hosts the ML Model data (e.g RAN Node, 0AM, OTT Server or core network (e.g., LMF or NWDAF). In various embodiments, the node hosting the ML Model data may trigger the setup of necessary resources to transfer the ML Model data, as described herein.

[0097] In various embodiments, the ML Model Data may include one or more of the following: the ML model / algorithm itself, in open or proprietary format, configuration parameters for an existing ML model in the UE, such as parameters for specific parts / layers / block of the model (like weights and biases, etc.), a hyperparameter for an existing ML model in the UE e.g., configuration for the pre / post processing of the input / output data features (like filtering, sampling, masking, etc.), and / or context description parameters needed at the UE to be able to determine when / if its ML model needs to be finetuned / updated e.g. network configurations or assistance information (like antenna panel configuration, cell / beam layout, movement paths, training data type availability / requirements, etc.).

[0098] Described herein related to the NW Initiated Resource Management to enable ML model Data Transfer in various embodiments is where the ML Model data is hosted in the RAN node and RAN Node identifies the need to transfer the data to one or more UE(s).

[0099] Accordingly, in various embodiments described herein is a method by which RAN as an entity hosting the ML Model Data can manage the setup / release of resources through enhanced interactions (NAS and NGAP protocols) with Core Network and UE. A method is described herein by which RAN Node as a node hosting ML Model Data (e.g, ML Model) can trigger the setup of the necessary resources to download the ML Model Data to UE(s).

[0100] In various embodiments, a RAN Node can indicate to the Core Network the need for transferring ML Model Data to UE which in turn enables the Core Network to trigger the setup of QoS Flows / PDU Sessions associated with this ML Model Data Transfer.

[0101] FIG. 3 is a diagram of an example interaction 300 between a UE, a RAN and a Core Network, according to one illustrated aspect of the disclosure. In various embodiments, the components shown in FIG. 3 may be similar to those described above in FIGS. 1 and 2 (e.g., UE 150, RAN 225 and core network 210).

[0102] In various embodiments, as shown in FIG. 3, the ML Model Data is shown as being resident in the RAN, for example. However, the ML Model Data may be resident in any network entity. At operation 301, the RAN transmits an NG application protocol (NGAP) message trigger to Core to setup or release ML Model Data associated QoS flow or PDU session. Upon receiving the trigger, the core can add ML Model Data associated QoS flows to existing PDU sessions or can also set up new ML-Data associated PDU Sessions.

[0103] At operation 302, the core sends a network access stratum (NAS) NW management of ML-Data associated QoS Flow / PDU Session message to the UE. In various embodiments, the message includes the NAS impact to indicate a specific cause in PDU Session Setup Request and this cause value is propagated to the RAN.

[0104] At operation 303, an NGAP message is transmitted by the core to the RAN. In various embodiments, the NGAP message is an ML Model Data associated QoS Flow / PDU Session management related interaction between Core Network and RAN message. In various embodiments, the core network initiates a setup of new QoS Flows associated with the ML Model Data Transfer, and includes an additional “Binding ID” which associates the ML Model Data and QoS Flow ID. In various embodiments, the core network may bypass NG-U Tunnel setup.

[0105] In various embodiments, the RAN node hosts the ML Model Data. In this case, the RAN triggers the ML Model Data transfer to the applicable UE(s). The RAN node initiated trigger is UE-associated and QoS Flows for ML Model Data associated data transfer can be initiated at any time while the UE is RRC-Connected. After a UE is RRC-Connected, the RAN Node checks if there is any ML Model Data to be delivered to the UE. It then sends a new message to the Core Network which includes the ML Model Data ID.

[0106] In various embodiments, the RAN may treat ML Model Data associated QoS Flows / PDU sessions differently compared to normal QoS Flows / PDU Sessions by being able to identify and pre-empt ML-Data associated QoS Flows / PDU Sessions.

[0107] FIG. 4 is a diagram of an example embodiment of signals and operations for a QoS flow setup among a UE, RAN and Core Network, according to one illustrated aspect of thedisclosure. In various embodiments, the components depicted in FIG. 4 may correspond to similar components described above in FIGS. 1 -3. It will be understood that a described signal may have associated operations and a described operation may have associated signals.

[0108] As mentioned above, when the ML Model Data is hosted in the RAN node itself, the RAN node needs to be able to identify which QoS flow (and the corresponding DRB) corresponds to the ML Model Data Flow. In various embodiments, such an identification is needed for subsequent release when the data transfer is complete or for pre-emption.

[0109] At operation 401, the RAN has the ML Model Data. Accordingly, at operation 402, the RAN node assigns the ML Model Data flow ID.

[0110] At operation 403, the UE connects to the RAN node. In various embodiments, the connection procedure may be in accordance to connection procedures known to persons of skill in the art.

[0111] At operation 404, the RAN identifies that the UE is a candidate for receiving the ML Model Data. In various embodiments, the RAN maintains the list of the ML Model Data that is available. When the UE connects, the RAN Node checks if the UE is a candidate to receive any of the available ML Model data by checking the UE's ML Capabilities and / or the ML functionality that's activated in the UE / RAN.

[0112] Accordingly, at operation 405, the RAN transmits an ML Model Data flow associated QoS flow resource setup notification message to the core and the core receives the ML Model Data flow associated QoS flow resource setup notification message. In various embodiments, the ML Model Data flow associated QoS flow resource setup notification message may include a UE- NGAP-ID-Pair, ML Model Data flow ID, notification cause (e.g., setup), and ML data flow description.

[0113] At operation 406, the core transmits an ML Model Data flow associated QoS flow resource setup request message to the RAN and the RAN receives the ML Model Data flow associated QoS flow resource setup request message. In various embodiments, the ML Model Data flow associated QoS flow resource setup request message includes the UE-NGAP-ID-Pair, ML Model Data flow resource binding, and resource preference information.

[0114] At operation 407, the RAN prepares a ML Model Data flow to QoS flow mapping table.

[0115] At operation 408, the UE and RAN perform an RRC reconfiguration that includes a DRB-ToAddMod. At operation 409, the RAN transmits an ML Model Data flow associated QoS flow resource setup response message to the core and the core receives the ML Model Data flow associated QoS flow resource setup response message. In various embodiments, the ML Model Data flow associated QoS flow resource setup response message includes the UE-NGAP-ID-Pair information.

[0116] Accordingly, at operation 410, the ML Model Data is transferred from the RAN node to the UE.

[0117] In various embodiments, where the RAN architecture is split-architecture based (e.g., the architecture is split over more than one CU-UP), additional signaling may be employed (e.g., over the El AP interface) between the CU-CP and CU-UP. For example, once the ML Model Data is made available in a CU-UP, additional signaling from the CU-UP may be utilized to the CU-CP indicating the availability of new ML Model Data in the CU-UP. This enables the CU-CP to check the eligible RRC-Connected UE(s) to which the ML Model Data needs to be transferred. In various embodiments, for example, this signaling is non-UE associated and CU-CP checks the eligible UE(s), or eligibility could also be checked by CU-UP per UE when the bearer context is setup in CU-UP and then notified to CU-CP.

[0118] Also, in various embodiments, the CU-CP may signal the CU-UP to start the transfer of the ML Model Data from the CU-UP to the UE. ML Model Data transfer is handled by the CU- UP (e.g., the data is transferred from the CU-UP to DU over the Fl-U Tunnel and then subsequently sent over to specific UE(s) using the corresponding DRB). The CU-CP sends this signal to the CU-UP after the necessary resource configuration / setup in RAN and UE are completed.

[0119] The operations of FIG. 4 are merely illustrative, and variations are contemplated to be within the scope of the present disclosure. In embodiments, the operations may include other operations not illustrated in FIG. 4. In embodiments, the operations may not include every operation illustrated in FIG. 4. In embodiments, the operations may be implemented in a different order than that illustrated in FIG. 4. Such and other embodiments are contemplated to be within the scope of the present disclosure. Persons of skill in the art will appreciate that, although various example components are described as perform various functions, other components may perform those functions described in FIG. 4.

[0120] As mentioned above, an ML Model Data associated PDU session or QoS flow release may be effected. For example, in various embodiments, when the ML Model Data is hosted in the RAN node, RAN node starts the ML Model Data Transfer to the UE after the QoS Flow (and the corresponding DRB) is configured. In various embodiments, the RAN keeps track of the completion of the ML Model Data Transfer. Hence, the RAN node identifies the corresponding QoS Flow using the mapping table created (e.g., operation 407 above) and initiates the release of the ML Model Data Transfer associated QoS Flow after the completion of the ML Model Data Transfer.

[0121] FIG. 5 is a diagram of an example embodiment of signals and operations for a QoS flow release among a UE, RAN and Core Network, according to one illustrated aspect of the disclosure. In various embodiments, the components depicted in FIG. 5 may correspond to similar components described above in FIGS. 1 -3. It will be understood that a described signal may have associated operations and a described operation may have associated signals.

[0122] At operation 501, the UE is RRC connected, a ML Model Data flow associated QoS flow is setup, and ML Model Data transfer is ongoing.

[0123] At operation 502, the RAN identifies that the ML Model Data transfer is complete. At operation 503, the RAN identifies the corresponding QoS flow ID using the mapping table. At operation 504, an RRC reconfiguration procedure occurs between theUE and the RAN (e.g., DRB- ToRelease).

[0124] Accordingly, at operation 505, the RAN transmits a PDU session resource notify message to the core and the core receives the PDU session resource notify message. In various embodiments, the PDU session resource notify message includes a PDU session notify transfer item ID: PDU session / QoS flow for ML Model Data transfer IE.

[0125] At operation 506, the UE, RAN and core perform a core initiated QoS flow release. At operation 507, the RAN determines if the ML Model Data transfer is complete.

[0126] In various embodiments, at operation 508, the RAN unassigns the ML Data Model flow ID.

[0127] In various embodiments, depending on the network and traffic conditions, the RAN Node may want to prioritize the normal QoS Flows (DRBs) and hence pre-empt ML Model Data associated QoS Flows. In such a scenario, the RAN node may decide to pre-empt MLModel Data associated QoS Flow (corresponding DRBs) before the ML Model Data Transfer is completed.

[0128] Similar to above, in a split architecture, additional signaling may be employed from CU-UP to CU-CP to indicate the completion of the transfer. In various embodiments, the ML Model Data transfer is handled by a first CU-UP (e.g., the data is transferred from the first CU-UP to DU over the Fl-U Tunnel and then subsequently sent over to specific UE(s) using the corresponding DRB). After the data transfer is completed, the first CU-UP notifies this to a second CU-CP so that the release of the ML Model Data associated QoS Flow / PDU Sessions can be triggered. The second CU-CP may then request release of resources used for the transfer.

[0129] The operations of FIG. 5 are merely illustrative, and variations are contemplated to be within the scope of the present disclosure. In embodiments, the operations may include other operations not illustrated in FIG. 5. In embodiments, the operations may not include every operation illustrated in FIG. 5. In embodiments, the operations may be implemented in a different order than that illustrated in FIG. 5. Such and other embodiments are contemplated to be within the scope of the present disclosure. Persons of skill in the art will appreciate that, although various example components are described as perform various functions, other components may perform those functions described in FIG. 5.

[0130] In various embodiments, additional information elements may be utilized as part of the signaling and operations described above and below.

[0131] For example, an ML Model Data Flow ID IE may be utilized to identify the ML Model Data type to be transferred to the UE. In various embodiments, the RAN node assigns a ML Model Data ID for each unique ML Model Data that needs to be transferred to the UE. For example, if there are two different ML Model Data types that need to be transferred to a UE, the RAN node assigns a separate unique ML Model Data ID for each ML Model Data. ML Model Data Flow ID assignment can be done as soon as the ML Model Data is available in the RAN node (non-split) or CU-UP (split). In split architecture, ML Model Data Flow ID assignment can be handled by either a first CU-UP or a second CU-CP.

[0132] An ML Model Data Flow Description IE may include the ML Model Data Type and ML Model Data Size. In various embodiments, the ML Model Data Type can either be ML Model or other ML Model Data such as Configuration parameters, hyper parameter or context descriptionparameters. ML Model Data size indicates the amount of data to be transferred.

[0133] An ML Model Data Flow to Resource Binding IE may be utilized where the core network initiates the creation of the necessary resources to enable the transfer of ML Model Data (the resource can be a new PDU session itself or a new QoS Flow within an existing PDU session). The association between the ML Model Data ID and the corresponding QoS Flow ID is indicated by the Core to the RAN by including the ML Model Data ID earlier provided by the RAN. The association between ML Model Data and QoS Flow ID can either be 1: 1 (a separate QoS Flow for each ML Model Data Flow) or N: 1 (single QoS Flow for multiple ML Model Data Flows). A New QoS Flow can be added to an existing PDU Session or a new PDU Session can be created. These can be controlled / selected through operator configurable parameter in the core network.

[0134] In various embodiments, the core network may include an ML Resource Configuration Preference indicating other configuration preferences specific to the resources that are associated with ML Model Data when setting up the QoS flow to transfer the ML Model data . For example, the core network can indicate whether the NG-U is tunnel is configured or bypassed. If the NG-U is bypassed, the RAN also does not create the tunnel endpoints. In various embodiments, the RAN Node may indicate its preference on NG-U tunnel setup / bypass and the core can make the final decision.

[0135] In various embodiments, a UE associated procedure to the SMF may be employed. FIG. 6 is a diagram of an example embodiment of signals and operations for a RAN initiated ML Model Data associated resource setup trigger among a RAN and Core Network, according to one illustrated aspect of the disclosure. In various embodiments, the components depicted in FIG. 6 may correspond to similar components described above in FIGS. 1-3. It will be understood that a described signal may have associated operations and a described operation may have associated signals

[0136] At operation 601, the RAN transmits an ML Model Data flow associated QoS flow resource setup notification to the core and the core receives the ML Model Data flow associated QoS flow resource setup notification. In various embodiments, the ML Model Data flow associated QoS flow resource setup notification includes the UE-NGAP-ID Pair, and an ML Model Data flow notify item. In various embodiments, the ML Model Data flow associated QoS flow resource setupnotification may be terminated at the SMF, or another network function, in the core network. The SMF, or other network function, then sets up new QoS Flow for the PDU Session.

[0137] In various embodiments, the ML Model Data flow associated QoS flow resource setup notification may include an IE “ML Model Data Flow Notify List”. For each ML Model Data to be transferred to the UE(s), the RAN includes ML Model Data Flow Notify Item. Each ML Model Data Flow Notify Items includes ML Model Data Flow ID (unique ID for each ML Model Data to be transferred), Notification Cause (whether to setup or release) and optionally ML Model Data Flow Parameter Set Index (with additional information on the ML Model Data).

[0138] The table below illustrates an example IE associated with the ML data flow associatedQoS flow resource setup notification.

[0139] The operations of FIG. 6 are merely illustrative, and variations are contemplated to be within the scope of the present disclosure. In embodiments, the operations may include other operations not illustrated in FIG. 6. In embodiments, the operations may not include every operation illustrated in FIG. 6. In embodiments, the operations may be implemented in a different order than that illustrated in FIG. 6. Such and other embodiments are contemplated to be within the scope of the present disclosure. Persons of skill in the art will appreciate that, although various example components are described as perform various functions, other components may perform those functions described in FIG. 6.

[0140] In various embodiments, an IE may be introduced in a PDU Session Resource Notify message or a PDU Session Resource Modify Indication message.

[0141] FIG. 7 is a diagram of an example embodiment of signals and operations for a RAN initiated ML Model Data associated resource setup trigger among a RAN and Core Network, according to another illustrated aspect of the disclosure. In various embodiments, the components depicted in FIG. 7 may correspond to similar components described above in FIGS. 1-3. It will be understood that a described signal may have associated operations and a described operation may have associated signals.

[0142] At operation 701, the RAN transmits a PDU session resource notify message to the core and the core receives the PDU session resource notify message. In various embodiments, the PDU session resource notify message includes the UE-NGAP-ID Pair, and a PDU session notify transfer item IE: ML Model Data flow notify item. In various embodiments, the RAN transmits this notification to the core (e.g., SMF) upon identifying an UE eligible to receive ML Model data , and the SMF sets up a new QoS Flow for the PDU Session.

[0143] The operations of FIG. 7 are merely illustrative, and variations are contemplated to be within the scope of the present disclosure. In embodiments, the operations may include other operations not illustrated in FIG. 7. In embodiments, the operations may not include every operation illustrated in FIG. 7. In embodiments, the operations may be implemented in a different order than that illustrated in FIG. 7. Such and other embodiments are contemplated to be within the scope of the present disclosure. Persons of skill in the art will appreciate that, although various example components are described as perform various functions, other components may perform those functions described in FIG. 7.

[0144] FIG. 8 is a diagram of an example embodiment of signals and operations for a RAN initiated ML Model Data associated resource setup trigger among a RAN and Core Network, according to another illustrated aspect of the disclosure. In various embodiments, the components depicted in FIG. 8 may correspond to similar components described above in FIGS. 1-3. It will be understood that a described signal may have associated operations and a described operation may have associated signals.

[0145] At operation 801, the RAN transmits a PDU session resource modify indication message to the core and the core receives the PDU session resource modify indication message. In various embodiments, the PDU session resource modify indication message 1includes the UE-NGAP-ID Pair, and a PDU session resource modify indication item IE: ML Model Data flow notify item. At operation 802, the core transmits a PDU session resource modify confirm message to the RAN and the RAN receives the PDU session resource modify confirm message.

[0146] The operations of FIG. 8 are merely illustrative, and variations are contemplated to be within the scope of the present disclosure. In embodiments, the operations may include other operations not illustrated in FIG. 8. In embodiments, the operations may not include every operation illustrated in FIG. 8. In embodiments, the operations may be implemented in a different order than that illustrated in FIG. 8. Such and other embodiments are contemplated to be within the scope of the present disclosure. Persons of skill in the art will appreciate that, although various example components are described as perform various functions, other components may perform those functions described in FIG. 8.

[0147] In various embodiments, a an NGAP UE Context Modification Indication procedure to AMF may be utilized.

[0148] FIG. 9 is a diagram of an example embodiment of signals and operations for a RAN initiated ML Model Data associated resource setup trigger among a RAN and Core Network, according to another illustrated aspect of the disclosure. In various embodiments, the components depicted in FIG. 9 may correspond to similar components described above in FIGS. 1-3. It will be understood that a described signal may have associated operations and a described operation may have associated signals.

[0149] At operation 901, the RAN transmits a UE context modification indication message to the core and the core receives the UE context modification indication message. In various embodiments, the UE context modification indication message includes the UE-NGAP-ID Pair, and an ML Model Data flow item. In various embodiments, the UE context modification indication message includes a ML Model Data flow notify list IE.

[0150] At operation 902, the core transmits a UE context modification confirm message to the RAN and the RAN receives the UE context modification confirm message.

[0151] In various embodiments, when the RAN decides to start the ML Model Data download to the UE, it sends UE context modification indication message to the AMF. Upon receiving the UE context modification indication message, the AMF triggers with PDU Session Setup towardsthe SMF including the same indication. The SMF then sets up a new QoS Flow for the PDU Session.

[0152] The operations of FIG. 9 are merely illustrative, and variations are contemplated to be within the scope of the present disclosure. In embodiments, the operations may include other operations not illustrated in FIG. 9. In embodiments, the operations may not include every operation illustrated in FIG. 9. In embodiments, the operations may be implemented in a different order than that illustrated in FIG. 9. Such and other embodiments are contemplated to be within the scope of the present disclosure. Persons of skill in the art will appreciate that, although various example components are described as perform various functions, other components may perform those functions described in FIG. 9.

[0153] FIG. 10 is a diagram of an example embodiment of signals and operations for a Core initiated ML Model Data flow associated QoS flow setup among a RAN and Core Network, according to another illustrated aspect of the disclosure. In various embodiments, the components depicted in FIG. 10 may correspond to similar components described above in FIGS. 1-3. It will be understood that a described signal may have associated operations and a described operation may have associated signals.

[0154] In various embodiments, the SMF may initiate the QoS Flow setup, and include the corresponding ML Model Data Flow ID. This enables the RAN to identify / map the ML Model Data Flow ID and the corresponding QoS Flow ID (and DRB).

[0155] At operation 1001, the core transmits a PDU Session Resource Setup / modify request message to the RAN and the RAN receives the PDU Session Resource Setup / modify request message. In various embodiments, the PDU Session Resource Setup / modify request message includes the UE-NGAP-ID Pair, and the PDU session resource modify request item for ML Model Data transfer.

[0156] At operation 1002, the RAN transmits as PDU session resource modify response to the core and the core receives the PDU session resource modify response.

[0157] The tables below illustrates example IES associated with the PDU Session Resource Setup / modify request message.

[0158] The operations of FIG. 10 are merely illustrative, and variations are contemplated to be within the scope of the present disclosure. In embodiments, the operations may include other operations not illustrated in FIG. 10. In embodiments, the operations may not include every operation illustrated in FIG. 10. In embodiments, the operations may be implemented in a different order than that illustrated in FIG. 10. Such and other embodiments are contemplated to be within the scope of the present disclosure. Persons of skill in the art will appreciate that, although various example components are described as perform various functions, other components may perform those functions described in FIG. 10.

[0159] FIG. 11 is a diagram of an example embodiment of signals and operations for end to end ML Model Data transfer among a UE, RAN and Core Network, according to one illustrated aspect of the disclosure. In various embodiments, the components depicted in FIG. 10 may correspond to similar components described above in FIGS. 1-3. It will be understood that a described signal may have associated operations and a described operation may have associated signals.

[0160] At operation 1101, the RAN has ML Model Data. In various embodiments, the RAN hosts the ML Model Data (e.g trained ML Model), and assigns a unique ML Model Data Flow ID to each unique ML Model Data.

[0161] Accordingly, at operation 1102, the UE connects to the RAN node. At operation 1103, the RAN identifies that the UE is a candidate for receiving ML Model Data.

[0162] At operation 1104, the RAN transmits a PDU session resource notify message to the core and the core receives the PDU session resource notify message. In various embodiments, the PDU session resource notify message includes the UE-NGAP-ID-Pair, ML Model Data flow ID, notification cause=setup, and ML Model Data flow description.

[0163] At operation 1105, the core transmits a PDU Session Resource Setup / modify request message to the RAN and the RAN receives the PDU Session Resource Setup / modify request message. In various embodiments, the PDU Session Resource Setup / modify request message includes the UE-NGAP-ID Pair, the PDU session resource modify request item for ML Model Data transfer.

[0164] At operation 1106, the RRC reconfiguration (DRB-ToAddMod) is performed between the UE and the RAN.

[0165] At operation 1107, the RAN transmits a PDU session resource modify response message to the core and the core receives the PDU session resource modify response message.

[0166] At operation 1108, the RAN sends the ML Model Data to the UE. At operation 1109, the RAN identifies that the Model Data transfer is complete.

[0167] At operation 1110, the RAN transmits a PDU session resource notify message to the core and core receives the PDU session resource notify message. In various embodiments, the PDU session resource notify message includes the UE-NGAP-ID-Pair, ML Model Data flow ID, notification cause=release.

[0168] At operation 1111, the core transmits a PDU session resource release command to the RAN and the RAN receives the PDU session resource release command. Accordingly, at operation 1112, the UE and the RAN perform an RRC reconfiguration (DRB-ToRelease).

[0169] At operation 1113, the RAN transmits a PDU session resource release response message to the core and core receives the PDU session resource release response message.

[0170] Accordingly, as described in FIG. 11, when the RAN Node hosts ML Model Data (e.g trained ML Model), the RAN Node assigns a unique ML Model Data Flow ID to each unique ML Model Data.

[0171] After the UE connects to the RAN node and the UE moves to RRC- Connected state, the RAN node checks whether the UE is eligible to receive any of the ML Model Data based onthe AIML Capabilities of the UE and availability of relevant ML Model Data in the RAN node. The RAN node identifies the corresponding list of ML Model Data Flow IDs for each eligible UE. The RAN node sends an NGAP: PDU Session Resource Notify message including the list of ML Model Data Flow ID, corresponding description and Notification cause set to “setup”.

[0172] The core network (e.g., SME) triggers the creation of necessary QoS Flows and / or PDU Sessions. The core network maps the list of ML Model Data Flows associated with the QoS Flows and / or PDU Sessions. This mapping (e.g. ML Model data flow ID information element) is sent from the core network to the RAN node in the NGAP: PDU Session Resource Setup / Modify Request.

[0173] The RAN node creates the mapping table entry for the UE with the mapping between ML Model Data Flows and the QoS Flows / PDU Sessions. The RAN node sets up the necessary UP resources including the Fl-U tunnels and also reconfigures the UE with the DRB(s) for each QoS Flow. The RAN Node sends NGAP: PDU Session Resource Setup / Modify Response. The RAN node starts the ML Model Data Transfer using the corresponding DRB.

[0174] In case ML Model Data Flow: QoS Flow: DRB mapping is not 1:1 :1, then packet identification / demarcation may be employed.

[0175] The RAN node monitors the ML Model Data Transfer and identifies whether the ML Model Data transfer is complete. In various embodiments, the RAN node decides to pre-empt the ML Model Data transfer due to enabling the prioritization of other DRB(s). The RAN node then sends a NGAP: PDU Session Resource Notify message including the QoS Flow ID / PDU Session ID together with the Notification Cause set to “Release”. It may also include an additional cause value set to pre-emption.

[0176] The core Network initiates the release of the QoS Flow ID / PDU Session ID by sending the NGAP: PDU session resource release command.

[0177] The RAN node initiates the release of all allocated resources including the Fl-U Tunnels and also reconfigures the UE to release the DRB. RAN Node sends the NGAP: PDU Session Resource Release Response to the Core Network. In various embodiments, the RAN Node keeps track of the ML Model Data Transfer status whether i pending or completed. If the status is pending, the RAN node may re-attempt the ML Model Data Transfer if the UE is still in RRC- Connected or in a subsequent RRC Connection after UE moves to RRC-Idle

[0178] The operations of FIG. 11 are merely illustrative, and variations are contemplated to be within the scope of the present disclosure. In embodiments, the operations may include other operations not illustrated in FIG. 11. In embodiments, the operations may not include every operation illustrated in FIG. 11. In embodiments, the operations may be implemented in a different order than that illustrated in FIG. 11. Such and other embodiments are contemplated to be within the scope of the present disclosure. Persons of skill in the art will appreciate that, although various example components are described as perform various functions, other components may perform those functions described in FIG. 11.

[0179] In various embodiments, in the case of a ML Model Data Transfer with the Model hosted in RAN, although the DRB (Data Radio Bearer) and any tunnels within RAN (Fl-U) in case of split architecture may be setup, there may not be a requirement for the NG-U tunnel.

[0180] Accordingly, in various embodiments, when the core sets up the resources for ML Model Data associated QoS Flow, it may bypass the Core-side NG-U Tunnel resources and not include those corresponding IE(s) to the RAN. The RAN also can bypass the RAN-side NG-U Tunnel resources. This may include differentiated handling at RAN and Core.

[0181] In various embodiments, the core can treat the ML Model -data associated QoS Flows the same as the normal QoS Flow. Both the Core and RAN can allocate and setup NG-U Tunnels even if they are not used during the data transfer. This enables a common handling irrespective of whether the resources are being set up for normal data transfer or ML Model associated data transfer.

[0182] In various embodiments, for a normal QoS Flow, release of the resources may be triggered by the entity originating the data transfer (e.g., the UE in case of Mobile Originated voice call or the Application Server in case of MT data call) and may be transparent to the RAN node. However, in case of ML Model -data associated data transfer, a CU-UP may monitor the completion of data transfer and notify a CU-CP to initiate the release of the resources.

[0183] In various embodiments, the ML Model data is hosted in the CU-CP itself, CU-CP itself monitor the completion of data transfer.

[0184] The following describes operations from the perspective of a network apparatus (e.g., first apparatus). From such a perspective, a method may include identifying, by a first apparatus hosting a machine learning (ML) Model Data, that a User Equipment (UE) in a radio resource control (RRC) connected state is a candidate for receiving the ML Model data;transmitting, by the first apparatus, based on the identifying, a first trigger message to a second apparatus to setup at least one of a dedicated protocol data unit (PDU) session or a dedicated quality of service (QoS) flow for the UE; receiving, by the first apparatus, from the second apparatus, a PDU session setup request or a QoS flow setup request; setting up, by the first apparatus, the data radio bearer associated with the PDU session or QoS flow; and transmitting, by the first apparatus, to the UE the ML Model Data over the data radio bearer based upon a determination to deliver the ML Model Data to the UE.

[0185] The following describes operations from the perspective of core network entity (e.g., first core network entity). From such a perspective, a method may include receiving, by a first core network entity, from a network access node, a first trigger message to setup at least one of a dedicated protocol data unit (PDU) session or a dedicated quality of service (QoS) flow; and transmitting, by the first core network entity, a PDU session setup request or a QoS flow setup request to the network access node.

[0186] The following describes operations from the perspective of a UE. From such a perspective, a method may include receiving, by the UE from a first apparatus, a protocol data unit (PDU) Session setup request or quality of service (QoS) flow setup request comprising an indication that the PDU session setup request or the QoS flow setup request is for delivery of a machine learning (ML) Model Data; and storing, by the UE, data received over a radio bearer associated with the PDU session or the QoS flow as data of an ML model.

[0187] FIG. 12 is a block diagram of a wireless station or node (e.g., UE, user device, AP, BS, eNB, gNB, RAN node, network node, TRP, or other node) 1200, according to one illustrated aspect of the present disclosure. The wireless station 1200 may include, for example, one or more (e.g., two as shown in FIG. 12) RF (radio frequency) or wireless transceivers 1202A, 1202B, where each wireless transceiver includes a transmitter to transmit signals and a receiver to receive signals. The wireless station also includes a processor or control unit / entity (controller) 1204 to execute instructions or software and control transmission and receptions of signals, and a memory 1206 to store data and / or instructions.

[0188] Processor 1204 may also make decisions or determinations, generate frames, packets or messages for transmission, decode received frames or messages for further processing, and other tasks or functions described herein. Processor 1204, which may be a baseband processor, for example, may generate messages, packets, frames or other signals for transmission via wirelesstransceiver 602 (1202A or 1202B). Processor 1204 may control transmission of signals or messages over a wireless network, and may control the reception of signals or messages, etc., via a wireless network (e.g., after being down- converted by wireless transceiver 602, for example). Processor 1204 may be programmable and capable of executing software or other instructions stored in memory or on other computer media to perform the various tasks and functions described above, such as one or more of the tasks or methods described above. Processor 1204 may be (or may include), for example, hardware, programmable logic, a programmable processor that executes software or firmware, and / or any combination of these. Using other terminology, processor 1204 and transceiver 602 together may be considered as a wireless transmitter / receiver system, for example.

[0189] In addition, referring to FIG. 12, a controller (or processor) 1208 may execute software and instructions, and may provide overall control for the station 1200, and may provide control for other systems not shown in FIG. 12, such as controlling input / output devices (e.g., display, keypad), and / or may execute software for one or more applications that may be provided on wireless station 1200, such as, for example, an email program, audio / video applications, a word processor, a Voice over IP application, or other application or software.

[0190] In addition, a storage medium may be provided that includes stored instructions, which when executed by a controller or processor may result in the processor 1204, or other controller or processor, performing one or more of the functions or tasks described above.

[0191] According to another example embodiment, RF or wireless transceiver(s) 1202A / 1202B may receive signals or data and / or transmit or send signals or data. Processor 1204 (and possibly transceivers 1202A / 1202B) may control the RF or wireless transceiver 1202A or 1202B to receive, send, broadcast or transmit signals or data.

[0192] Example embodiments are provided or described for each of the example methods, including: An apparatus (e.g., 1200, FIG. 12) including means (e.g., processor 1204, RF transceivers 1202A and / or 1202B, and / or memory 1206, in FIG. 12) for carrying out any of the methods; a non-transitory computer-readable storage medium (e.g., memory 1206, FIG. 12) comprising instructions stored thereon that, when executed by at least one processor (processor 1204, FIG. 12), are configured to cause a computing system (e.g., 1200, FIG. 12) to perform any of the example methods; and an apparatus (e.g., 1200, FIG. 12) including at least one processor (e.g., processor 1204, FIG. 12), and at least one memory (e.g., memory 1206, FIG. 12) includingcomputer program code, the at least one memory (1206) and the computer program code configured to, with the at least one processor (1204), cause the apparatus (e.g., 1200) at least to perform any of the example methods.

[0193] Further embodiments of the present disclosure include the following examples.

[0194] Example 1.1. An apparatus, comprising: means for identifying, by a first apparatus hosting a machine learning (ML) Model Data, that a User Equipment (UE) in a radio resource control (RRC) connected state is a candidate for receiving the ML Model data; means for transmitting, by the first apparatus, based on the identifying, a first trigger message to a second apparatus to setup at least one of a dedicated protocol data unit (PDU) session or a dedicated quality of service (QoS) flow for the UE; means for receiving, by the first apparatus, from the second apparatus, a PDU session setup request or a QoS flow setup request; means for setting up, by the first apparatus, a data radio bearer associated with the PDU session or QoS flow; and means for transmitting, by the first apparatus, to the UE the ML Model Data over the data radio bearer based upon a determination to deliver the ML Model Data to the UE.

[0195] Example 1.2. The apparatus of example 1.1, wherein the first trigger message includes an indication that the requested PDU session or QoS flow is for delivery of an ML Model data.

[0196] Example 1.3. The apparatus as in any preceding example, wherein the first trigger message includes an ML Model Data ID.

[0197] Example 1.4. The apparatus as in any preceding example, further comprising: means for including, by the first apparatus, in the first trigger message to the second apparatus a binding identifier (binding ID); means for receiving, by the first apparatus, from the second apparatus a PDU session setup request of QoS flow setup request including the binding identifier; means for determining, by the first apparatus, based on the binding identifier, a mapping between the received PDU session setup request or QoS flow setup request and the first trigger message; and means for deciding based on the determining, by the first apparatus, to transmits theidentified UE the ML Model Data over a data radio bearer associated with the PDU session or QoS flow.

[0198] Example 1.5. The apparatus as in any preceding example, wherein the binding ID is the ML Model Data ID.

[0199] Example 1.6. The apparatus as in any preceding example, further comprising: means for determining, by the first apparatus, that a PDU session setup request or QoS flow setup request received from the second apparatus is for delivery of an ML Model Data; and means for setting up, by the first apparatus, a data radio bearer corresponding to the requested PDU session or QoS flow without setting up an associated tunnel between the first apparatus and a third apparatus.

[0200] Example 1.7. The apparatus of example 1.6, wherein the determining that a PDU session setup request or QoS flow setup request received from the second apparatus is for delivery of an ML Model Data includes receiving in the PDU session setup request or the QoS flow setup request a binding ID or an ML Model Data ID.

[0201] Example 1.8. The apparatus of example 1.6 or 1.7, wherein the transmission to the second apparatus in response to the received PDU session setup request or QoS flow setup request includes at least one of: an indicator to not setup the tunnel between the first apparatus and the third apparatus, or not including a downlink tunnel endpoint for a tunnel between the first apparatus and the third apparatus.

[0202] Example 1.9. The apparatus as in any preceding example, further comprising: means for detecting, by the first apparatus, the end of delivery of the ML Model Data to the UE; means for transmitting, to the second apparatus, a notification indicating the end of delivery or a second trigger message to release the PDU session or to release the QoS flow; means for receiving, by the first apparatus, a PDU session release request of QoS flow release request from the second apparatus; and means for releasing the data radio bearer associated with the said PDU session or QoS flow.

[0203] Example 1.10. The apparatus of example 1.9, wherein the notification or the second trigger message includes at least one of the ML Model Data ID or the binding identifier.

[0204] Example 1.11. The apparatus of example 1.9 or 1.10 wherein the PDU sessionrelease request or the QoS flow release request includes at least one of the ML Model Data ID or the binding identifier.

[0205] Example 1.12. The apparatus as in any preceding example, further comprising: means for determining, by the first apparatus, an ML Model data size or an ML Model data type associated with the ML Model data; and means for identifying, by the first apparatus, the UE based on the determined ML Model data size or ML Model data type.

[0206] Example 1.13. The apparatus as in any preceding example, wherein the first apparatus is an NG-RAN node of a 5G cellular network.

[0207] Example 1.14. The apparatus as in any preceding example, wherein the first apparatus is a gNB-CU CP of a gNB split between a gNB-CU CP and one or more gNB-CU UP(s).

[0208] Example 1.15. The apparatus as in any preceding example, further comprising: means for receiving, from a gNB-CU UP, a notification of availability of ML Model data in the gNB-CU UP; and means for identifying a UE and sending the first trigger message based on receiving this notification.

[0209] Example 1.16. The apparatus of example 1.14 or 1.15 further comprising receiving from the gNB-CU UP at least one of an ML Model Data ID or an ML Model Data Size or an ML Model Data type.

[0210] Example 1.17. The apparatus of example 1.15 to 1.16 further comprising: means for sending to the gNB-CU UP a request to start the ML Model data delivery over an Fl-U tunnel associated with a context of the identified UE.

[0211] Example 1.18. The apparatus of example 1.14 to 1.17 further comprising: means for receiving from the gNB-CU UP an indication of end of ML Model Data delivery; and means for sending the notification of end delivery to the second apparatus or sending the second trigger message to the second apparatus based on the receiving of this indication from gNB-CU UP.

[0212] Example 1.19. The apparatus as in any preceding example, wherein the second apparatus is a core network entity and the core network entity is an Access Management Lunction (AMF) or a Session Management Function (SMF) of a 5G cellular network.

[0213] Example 1.20. The apparatus as in any preceding example, wherein the third apparatus is another core network entity and wherein the another core network entity is a user plane function (UPF) of a 5G cellular network.

[0214] Example 1.21. The apparatus as in any preceding example, wherein the first trigger message, the second trigger message and the notification message are NG Application Protocol (NGAP) messages.

[0215] Example 2.1. An apparatus, comprising: means for receiving, by a first core network entity, from a network access node, a first trigger message to setup at least one of a dedicated protocol data unit (PDU) session or a dedicated quality of service (QoS) flow; and means for transmitting, by the first core network entity, a PDU session setup request or a QoS flow setup request to the network access node.

[0216] Example 2.2. The apparatus of example 2.1, wherein the first trigger message includes an indication that the requested PDU session or QoS flow is for delivery of a machine learning (ML) Model data.

[0217] Example 2.3. The apparatus of example 2.1, wherein the first trigger message includes an ML Model Data ID.

[0218] Example 2.4. The apparatus as in any one of examples 2.1 to 2.3, further comprising: means for receiving, by the first core network entity, in the first trigger message from the network access node a binding identifier (binding ID); and means for transmitting, by the first core network entity, to the network access node, in response to the first trigger message, a PDU session setup request of QoS flow setup request including the binding ID.

[0219] Example 2.5. The apparatus as in any one of examples 2.1 to 2.4, wherein the binding ID is the ML Model Data ID.

[0220] Example 2.6. The apparatus as in any one of examples 2.1 to 2.5, further comprising: means for determining, by the first core network entity, from receiving the first trigger message, that the requested PDU session setup request or QoS flow setup request is for delivery of an ML Model Data; andmeans for transmitting, by the first core network entity, the PDU Session setup request or QoS flow setup request without an Uplink tunnel endpoint of a second core network entity.

[0221] Example 2.7. The apparatus as in any one of examples 2.1 to 2.6, further comprising means for including, in the PDU session setup request or the QoS flow setup request a Non Access Stratum (NAS) container towards the UE indicating that the PDU session setup or QoS flow setup is associated with an ML Model data delivery.

[0222] Example 2.8. The apparatus as in any one of examples 2.1 to 2.7, further comprising means for including, in the PDU session setup request or the QoS flow setup request a Non Access Stratum (NAS) container towards the UE indicating an ML Model Data ID.

[0223] Example 2.9. The apparatus as in any one of examples 2.1 to 2.8, further comprising: means for receiving, by the first core network entity from the network access node, a notification indicating the end of the ML Model Data delivery or a second trigger message to release the PDU session or release the QoS flow; and means for transmitting, by the first core network entity, to the network access node in response to the notification or second trigger message, a PDU session release request or QoS flow release request.

[0224] Example 2.10. The apparatus of example 2.9, wherein the notification or the second trigger message includes at least one of the ML Model Data ID or the binding identifier.

[0225] Example 2.11. The apparatus of example 2.8 or 2.10, wherein the PDU session release request or the QoS flow release request includes at least one of the ML Model Data ID or the binding identifier.

[0226] Example 2.12. The apparatus as in any one of examples 2.1 to 2.11, wherein the network access node is an NG-RAN node of a 5G cellular network.

[0227] Example 2.13. The apparatus as in any one of examples 2.1 to 2.12, the first core network entity is an Access Management Function (AMF) or a Session Management Function (SMF) of a 5G cellular network.

[0228] Example 2.14. The apparatus as in any one of examples 2.1 to 2.13, wherein the second core network entity is a user plane function (UPF) of a 5G cellular network.

[0229] Example 2.15. The apparatus as in any one of examples 2.1 to 2.14, wherein the first trigger message, the second trigger message and the notification message are NG ApplicationProtocol (NGAP) messages.

[0230] Example 3.1. A user equipment (UE), comprising: means for receiving, by the UE from a core network entity, a protocol data unit (PDU) Session setup request or quality of service (QoS) flow setup request comprising an indication that the PDU session setup request or the QoS flow setup request is for delivery of a machine learning (ML) Model Data; and means for storing, by the UE, data received over a radio bearer associated with the PDU session or the QoS flow as data of an ML model.

[0231] Example 3.2. The UE of example 3.1, wherein the indication is an ML Model Data ID.

[0232] Example 3.3. The UE of example 3.2, further comprising means for associating and storing, by the UE, the ML Model Data ID with the data received over the radio bearer associated with the PDU session or the QoS flow.

[0233] Example 3.4. The UE as in any one of examples 3.1 to 3.3, wherein the receiving an indication that the PDU session setup request or the QoS flow setup request is for delivery of an ML Model Data includes receiving the indication in a Non Access Stratum (NAS) message from a core network entity.

[0234] Example 3.5. The UE as in any one of examples 3.1 to 3.4, wherein the receiving an indication that the PDU session setup request or the QoS flow setup request is for delivery of an ML Model Data, comprises receiving the indication in a Radio Resource Control (RRC) message from a network access node.

[0235] Example 3.6. The UE of example 3.5, wherein the network access node is an NG- RAN node of a 5G cellular network.

[0236] Example 3.7. The UE as in any one of examples 3.1 to 3.6, wherein a core network entity is an Access Management Function (AMF) or a Session Management Function (SMF) of a 5G cellular network.

[0237] The embodiments and aspects disclosed herein are examples of the present disclosure and may be embodied in various forms. For instance, although certain embodiments herein are described as separate embodiments, each of the embodiments herein may be combined with one or more of the other embodiments herein. Specific structural and functional details disclosed herein are not to be interpreted as limiting, but as a basis for the claims and as a representative basis forteaching one skilled in the art to variously employ the present disclosure in virtually any appropriately detailed structure. Like reference numerals may refer to similar or identical elements throughout the description of the figures.

[0238] Although in various embodiments, protocols such as 5G protocols may be described, persons of skill in the art will understand that other protocols (e.g., 6G protocols) may be utilized for any of the operations and / or signaling described above along with their associated data, IES, messaging, or the like.

[0239] The phrases “in an aspect,” “in aspects,” “in various aspects,” “in some aspects,” or “in other aspects” may each refer to one or more of the same or different aspects in accordance with this present disclosure. The phrase “a plurality of’ may refer to two or more.

[0240] In various embodiments, the terms “first message” and “second message”, as well as any subsequent messages may refer to any messages that are transmitted or received in an order and are not necessarily limited to any particular message.

[0241] The phrases “in an embodiment,” “in embodiments,” “in various embodiments,” “in some embodiments,” or “in other embodiments” may each refer to one or more of the same or different embodiments in accordance with the present disclosure. A phrase in the form “A or B” means “(A), (B), or (A and B).” A phrase in the form “at least one of A, B, or C” means “(A); (B); (C); (A and B); (A and C); (B and C); or (A, B, and C) ”

[0242] Any of the herein described methods, programs, algorithms or codes may be converted to, or expressed in, a programming language or computer program. The terms “programming language” and “computer program,” as used herein, each include any language used to specify instructions to a computer, and include (but is not limited to) the following languages and their derivatives: Assembler, Basic, Batch files, BCPL, C, C+, C++, Delphi, Fortran, Java, JavaScript, machine code, operating system command languages, Pascal, Perl, PL1, Python, scripting languages, Visual Basic, metalanguages which themselves specify programs, and all first, second, third, fourth, fifth, or further generation computer languages. Also included are database and other data schemas, and any other meta- languages. No distinction is made between languages which are interpreted, compiled, or use both compiled and interpreted approaches. No distinction is made between compiled and source versions of a program. Thus, reference to a program, where the programming language could exist in more than one state (such as source, compiled, object, orlinked) is a reference to any and all such states. Reference to a program may encompass the actual instructions and / or the intent of those instructions.

[0243] While aspects of the present disclosure have been shown in the drawings, it is not intended that the present disclosure be limited thereto, as it is intended that the present disclosure be as broad in scope as the art will allow and that the specification be read likewise. Therefore, the above description should not be construed as limiting, but merely as exemplifications of particular aspects. Those skilled in the art will envision other modifications within the scope and spirit of the claims appended hereto.

Claims

WHAT IS CLAIMED IS:

1. A method, including: identifying, by a first apparatus hosting a machine learning (ML) Model Data, that a User Equipment (UE) in a radio resource control (RRC) connected state is a candidate for receiving the ML Model data; transmitting, by the first apparatus, based on the identifying, a first trigger message to a second apparatus to setup at least one of a dedicated protocol data unit (PDU) session or a dedicated quality of service (QoS) flow for the UE; receiving, by the first apparatus, from the second apparatus, a PDU session setup request or a QoS flow setup request; setting up, by the first apparatus, a data radio bearer associated with the PDU session or QoS flow; and transmitting, by the first apparatus, to the UE the ML Model Data over the data radio bearer based upon a determination to deliver the ML Model Data to the UE.

2. The method as claimed in claim 1, wherein the first trigger message includes an indication that the requested PDU session or QoS flow is for delivery of an ML Model data.

3. The method as claimed in any preceding claim, wherein the first trigger message includes an ML Model Data ID.

4. The method as claimed in any preceding claim, further comprising: including, by the first apparatus, in the first trigger message to the second apparatus a binding identifier (binding ID); receiving, by the first apparatus, from the second apparatus a PDU session setup request of QoS flow setup request including the binding identifier; determining, by the first apparatus, based on the binding identifier, a mapping between the received PDU session setup request or QoS flow setup request and the first trigger message; and deciding based on the determining, by the first apparatus, to transmits the identified UE the ML Model Data over a data radio bearer associated with the PDU session or QoS flow.

5. The method as claimed in any preceding claim, wherein the binding ID is the ML Model Data ID.

6. The method as claimed in any preceding claim, further comprising: determining, by the first apparatus, that a PDU session setup request or QoS flow setup request received from the second apparatus is for delivery of an ML Model Data; and setting up, by the first apparatus, a data radio bearer corresponding to the requested PDU session or QoS flow without setting up an associated tunnel between the first apparatus and a third apparatus.

7. The method as claimed in claim 6, wherein the determining that a PDU session setup request or QoS flow setup request received from the second apparatus is for delivery of an ML Model Data includes receiving in the PDU session setup request or the QoS flow setup request a binding ID or an ML Model Data ID.

8. The method as claimed in claim 6 or 7, wherein the transmission to the second apparatus in response to the received PDU session setup request or QoS flow setup request includes at least one of: an indicator to not setup the tunnel between the first apparatus and the third apparatus, or not including a downlink tunnel endpoint for a tunnel between the first apparatus and the third apparatus.

9. The method as claimed in any preceding claim, further comprising: detecting, by the first apparatus, the end of delivery of the ML Model Data to the UE; transmitting, to the second apparatus, a notification indicating the end of delivery or a second trigger message to release the PDU session or to release the QoS flow; receiving, by the first apparatus, a PDU session release request of QoS flow release request from the second apparatus; and releasing the data radio bearer associated with the said PDU session or QoS flow.4010. The method as claimed in claim 9, wherein the notification or the second trigger message includes at least one of the ML Model Data ID or the binding identifier.

11. The method as claimed in claim 9 or 10, wherein the PDU session release request or the QoS flow release request includes at least one of the ML Model Data ID or the binding identifier.

12. The method as claimed in any preceding claim, further comprising: determining, by the first apparatus, an ML Model data size or an ML Model data type associated with the ML Model data, wherein a Model data type includes one or more of a ML Model, configuration parameters, hyper parameter or context description parameters ; and identifying, by the first apparatus, the UE based on the determined ML Model data size or ML Model data type.

13. The method as claimed in any preceding claim, wherein the first apparatus is an NG-RAN node of a 5G cellular network.

14. The method as claimed in any preceding claim, wherein the first apparatus is a gNB-CU CP of a gNB split between a gNB-CU CP and one or more gNB-CU UP(s).

15. The method as claimed in claim 14, further comprising: receiving, from a gNB-CU UP, a notification of availability of ML Model data in the gNB- CU UP; and identifying a UE and sending the first trigger message based on receiving this notification.

16. The method as claimed in claim 14 or 15, further comprising receiving from the gNB-CU UP at least one of an ML Model Data ID or an ML Model Data Size or an ML Model Data type.

17. The method as claimed in any one of claims 15 to 16, further comprising:41sending to the gNB-CU UP a request to start the ML Model data delivery over an Fl-U tunnel associated with a context of the identified UE.

18. The method as claimed in any one of claims 14 to 17, further comprising: receiving from the gNB-CU UP an indication of end of ML Model Data delivery; and sending the notification of end delivery to the second apparatus or sending the second trigger message to the second apparatus based on the receiving of this indication from gNB-CU UP.

19. The method as claimed in any preceding claim, wherein the second apparatus is a core network entity and the core network entity is an Access Management Function (AMF) or a Session Management Function (SMF) of a 5G cellular network.

20. The method as claimed in any preceding claim, wherein the third apparatus is another core network entity and wherein the another core network entity is a user plane function (UPF) of a 5G cellular network.

21. The method as claimed in any preceding claim, wherein the first trigger message, the second trigger message and the notification message are NG Application Protocol (NGAP) messages.

22. A method, comprising: receiving, by a first core network entity, from a network access node, a first trigger message to setup at least one of a dedicated protocol data unit (PDU) session or a dedicated quality of service (QoS) flow; and transmitting, by the first core network entity, a PDU session setup request or a QoS flow setup request to the network access node.

23. The method as claimed in claim 22, wherein the first trigger message includes an indication that the requested PDU session or QoS flow is for delivery of a machine learning (ML) Model data.

24. The method as claimed in claim 22, wherein the first trigger message includes an ML Model Data ID.

25. The method as claimed in any one of claims 22 to 24, further comprising: receiving, by the first core network entity, in the first trigger message from the network access node a binding identifier (binding ID); and transmitting, by the first core network entity, to the network access node, in response to the first trigger message, a PDU session setup request of QoS flow setup request including the binding ID.

26. The method as claimed in any one of claims 22 to 25, wherein the binding ID is the ML Model Data ID.

27. The method as claimed in any one of claims 22 to 26, further comprising: determining, by the first core network entity, from receiving the first trigger message, that the requested PDU session setup request or QoS flow setup request is for delivery of an ML Model Data; and transmitting, by the first core network entity, the PDU Session setup request or QoS flow setup request without an Uplink tunnel endpoint of a second core network entity.

28. The method as claimed in any one of claims 22 to 27, further comprising including, in the PDU session setup request or the QoS flow setup request a Non Access Stratum (NAS) container towards the UE indicating that the PDU session setup or QoS flow setup is associated with an ML Model data delivery.

29. The method as claimed in any one of claims 22 to 28, further comprising including, in the PDU session setup request or the QoS flow setup request a Non Access Stratum (NAS) container towards the UE indicating an ML Model Data ID.

30. The method as claimed in any one of claims 22 to 29, further comprising:receiving, by the first core network entity from the network access node, a notification indicating the end of the ML Model Data delivery or a second trigger message to release the PDU session or release the QoS flow; and transmitting, by the first core network entity, to the network access node in response to the notification or second trigger message, a PDU session release request or QoS flow release request.

31. The method as claimed in claim 30, wherein the notification or the second trigger message includes at least one of the ML Model Data ID or the binding identifier.

32. The method as claimed in claim 30 or 31, wherein the PDU session release request or the QoS flow release request includes at least one of the ML Model Data ID or the binding identifier.

33. The method as claimed in any one of claims 22 to 32, wherein the network access node is an NG-RAN node of a 5G cellular network.

34. The method as claimed in any one of claims 22 to 33, wherein the first core network entity is an Access Management Function (AMF) or a Session Management Function (SMF) of a 5G cellular network.

35. The method as claimed in any one of claims 22 to 34, wherein the second core network entity is a user plane function (UPF) of a 5G cellular network.

36. The method as claimed in any one of claims 22 to 35, wherein the first trigger message, the second trigger message and the notification message are NG Application Protocol (NGAP) messages.

37. A method, comprising: receiving, by a user equipment (UE) from a core network entity, a protocol data unit (PDU) Session setup request or quality of service (QoS) flow setup request comprising an indication that44the PDU session setup request or the QoS flow setup request is for delivery of a machine learning (ML) Model Data; and storing, by the UE, data received over a radio bearer associated with the PDU session or the QoS flow as data of an ML model.

38. The method as claimed in claim 37, wherein the indication is an ML Model Data ID.

39. The method as claimed in claim 38, further comprising associating and storing, by the UE, the ML Model Data ID with the data received over the radio bearer associated with the PDU session or the QoS flow.

40. The method as claimed in any one of claims 37 to 39, wherein the receiving an indication that the PDU session setup request or the QoS flow setup request is for delivery of an ML Model Data comprises receiving the indication in a Non Access Stratum (NAS) message from a core network entity.

41. The method as claimed in any one of claims 37 to 40, wherein the receiving an indication that the PDU session setup request or the QoS flow setup request is for delivery of an ML Model Data, comprises receiving the indication in a Radio Resource Control (RRC) message from a network access node.

42. The method as claimed in claim 41, wherein the network access node is an NG- RAN node of a 5G cellular network.

43. The method as claimed in any one of claims 37 to 42, wherein the core network entity is an Access Management Function (AMF) or a Session Management Function (SMF) of a 5G cellular network.

44. An apparatus, comprising: at least one processor; and45at least one memory storing instructions which, when executed by the at least one processor, causes the apparatus at least to perform a method as in any one of claims 1-36.

45. A user equipment (UE), comprising: at least one processor; and at least one memory storing instructions which, when executed by the at least one processor, causes the UE at least to perform a method as in any one of claims 37-44.

46. A processor-readable medium storing instructions which, when executed by at least one processor of an apparatus, cause the apparatus at least to perform a method as in any one of claims 1 to 44.

47. An apparatus, comprising: at least one processor; and at least one memory storing instructions which, when executed by the at least one processor, causes the apparatus at least to perform: identifying, by a first apparatus hosting a machine learning (ML) Model Data, that a User Equipment (UE) in a radio resource control (RRC) connected state is a candidate for receiving the ML Model data; transmitting, by the first apparatus, based on the identifying, a first trigger message to a second apparatus to setup at least one of a dedicated protocol data unit (PDU) session or a dedicated quality of service (QoS) flow for the UE; receiving, by the first apparatus, from the second apparatus, a PDU session setup request or a QoS flow setup request; setting up, by the first apparatus, the data radio bearer associated with the PDU session or QoS flow; and transmitting, by the first apparatus, to the UE the ML Model Data over a data radio bearer based upon a determination to deliver the ML Model Data to the UE.

48. An apparatus, comprising: at least one processor; and46at least one memory storing instructions which, when executed by the at least one processor, causes the apparatus at least to perform: receiving, by a first core network entity, from a network access node, a first trigger message to setup at least one of a dedicated protocol data unit (PDU) session or a dedicated quality of service (QoS) flow; and transmitting, by the first core network entity, a PDU session setup request or a QoS flow setup request to the network access node.

49. A user equipment (UE), comprising: receiving, by the UE from a core network entity, a protocol data unit (PDU) Session setup request or quality of service (QoS) flow setup request comprising an indication that the PDU session setup request or the QoS flow setup request is for delivery of a machine learning (ML) Model Data; and storing, by the UE, data received over a radio bearer associated with the PDU session or the QoS flow as data of an ML model.47

Images