Managing node, executing node, requesting node and methods in a communications network

Abstract

A method performed by a managing node for training and / or managing a machine learning, ML, model associated with a vertical application layer, VAL, service in a communications network is provided The managing node manages (201) one or more operations in the communications network. The one or more operations are related to the ML model. The managing node executes (202) at least one out of the managed one or more operations in the communications network.

Description

[0001] MANAGING NODE, EXECUTING NODE, REQUESTING NODE AND METHODS IN A

[0002] COMMUNICATIONS NETWORK

[0003] TECHNICAL FIELD

[0004] Embodiments herein relate to a managing node, an execution node and a requesting node and methods therein. In some aspects, they relate to training and / or managing an machine learning model associated with a vertical application layer in a communications network.

[0005] BACKGROUND

[0006] In a typical wireless communication network, wireless devices, also known as wireless communication devices, mobile stations, stations (STA) and / or User Equipments (UE), communicate via a Wide Area Network or a Local Area Network such as a Wi-Fi network or a cellular network comprising a Radio Access Network (RAN) part and a Core Network (CN) part. The RAN covers a geographical area which is divided into service areas or cell areas, which may also be referred to as a beam or a beam group, with each service area or cell area being served by a radio network node such as a radio access node e.g., a Wi-Fi access point or a radio base station (RBS), which in some networks may also be denoted, for example, a NodeB, eNodeB (eNB), or gNB as denoted in Fifth Generation (5G) telecommunications. A service area or cell area is a geographical area where radio coverage is provided by the radio network node. The radio network node communicates over an air interface operating on radio frequencies with the wireless device within range of the radio network node.

[0007] 3GPP is the standardization body for specifying the standards for a cellular system evolution, e.g., including 3G, 4G, 5G and future evolutions. Specifications for the Evolved Packet System (EPS), also called a Fourth Generation (4G) network, have been completed within the 3rd Generation Partnership Project (3GPP). As a continued network evolution, new releases of 3GPP specify a 5G network also referred to as 5G New Radio (NR).

[0008] Multi-antenna techniques may significantly increase the data rates and reliability of a wireless communication system. The performance is in particular improved if both the transmitter and the receiver are equipped with multiple antennas, which results in a Multiple-Input Multiple-Output (MIMO) communication channel. Such systems and / or related techniques are commonly referred to as MIMO.

[0009] In addition to faster peak Internet connection speeds, 5G planning aims at higher capacity than current 4G, allowing higher number of mobile broadband users per area unit, and allowing consumption of higher or unlimited data quantities in gigabyte per month and user. This would make it feasible for a large portion of the population to stream high-definition media many hours per day with their mobile devices, when out of reach of Wi-Fi hotspots. 5G research and development also aims at improved support of machine to machine communication, also known as the Internet of things, aiming at lower cost, lower battery consumption and lower latency than 4G equipment.

[0010] Artificial Intelligence (Al) / Machine learning (ML) relevant topics have been studied in Release 18 SA2. For example, Federated Learning (FL) among multiple Network Data Analytics Functions (NWDAF) within 5G Core (5GC) has been studied in SA2 Release 18 work item eNA_Ph3 (3GPP TS 23.288 v18.3.0). 5GC supporting application layer AI / ML operations has been studied in SA2 Release 18 work item AIMLsys (3GPP TS 23.501 v18.3.0, 3GPP TS 23.502 v18.3.0). Recently, a new Release 19 SA6 Study Item Description (SID) was agreed on SA#101 plenary meeting. The new SA6 SID is about study on application layer support for AI / ML services, main content is as follows.

[0011] SA1 Rel-18 identified requirements (3GPP TS 22.261 v19.4.0) for the support of AI / ML model distribution, transfer, training for various applications, e.g., video / speech recognition, robot control, automotive, and initiated AIML_MT_Ph2 in Rel-19 for supporting Distributed Al training / inference based on direct device connection. Such use cases and requirements have application layer impacts.

[0012] SA6 has specified enablement services to ensure efficient use and deployment of vertical applications over 3GPP system. It is required to study SA1 requirements and identify any application enablement impacts and architecture requirements needed to provide assistance in AI / ML operations (model distribution, transfer and training) at the application layer.

[0013] In addition, SA6 has specified the application layer architecture to enable data analytics as a new Service Enabler Arcitecture Layer (SEAL) service, also known as Application Data Analytics Enabler Server (ADAES), in 3GPP TS 23.436 v18.1.0. Such architecture provides an application layer analytics framework which offers generic analytics exposure and value-added services for verticals and ASPs. This includes application layer analytics related to end to end application performance, edge load, service API availability, location accuracy and slice-related performance and fault analysis. In Rel-19, the ADAES functional architecture requires enhancements to further improve and enhance functionality in 3GPP for improved support for analytics using AI / ML methods.

[0014] In the proposed study, the focus will be on supporting application layer AI / ML enablement for:

[0015] 1. Addressing SA1 requirements and in particular support the application specific I VAL layer in the transfer / distribution / training covering both on-network and off- network models, taking into consideration system architecture capabilities developed by SA2 to assist AI / ML services.

[0016] 2. Enhancing and potentially extending existing analytics enablement services, as provided by SEAL ADAES, using AI / ML support services.

[0017] Possible new areas of study include:

[0018] - Study new or enhancement of existing SA6 enablers, e.g., ADAES, EDGEAPP, for supporting AI / ML service enablement, based on the SA1 requirements and SA2 capabilities. Part of the study will also investigate the enablement for device-to-device AI / ML model transfer / distribution and training as discussed in SA1 Rel-19 AIML_MT_Ph2.

[0019] - Study the implications when using AI / ML methods for ADAES analytics. In particular, to study enhanced or new enablement capabilities for supporting ML model training I inference and federated learning in coordinated ADAES deployments. Such support can be for example the assistance for the discovery and selection of the SA6 enablers, e.g., ADAEC, edge or central ADAES, to serve as ML training / inference entities or FL members in federated learning case.

[0020] - Enhancements of edge analytics service to support more advanced EDGEAPP scenarios, e.g., support the selection of common EAS.

[0021] - Possibly study whether and how digital twin modelling data at the DN side can be utilized as part of the AI / ML model lifecycle, at the application layer. In particular, whether the digital twin model output can be used as input for the ML model lifecycle operations, e.g., based on the use case in 3GPP TS 22.104 v.19.1.0 A.2.3.4, and how this impacts the SA6 existing or potentially new functionalities.

[0022] The SA6 objective of the study is to enable support for AI / ML services at application enablement layer, which include the following:

[0023] 1) Analyze Rel-18 and Rel-19 requirements in 3GPP TS 22.261 v.19.4.0 related to AI / ML model distribution, transfer, training and further identify key issues, develop corresponding architectural requirements at the application enablement layer and potential enhancements to the application layer architecture.

[0024] NOTE 1: This objective will leverage existing SA2 5GC capabilities developed in Re-18 AIM Lsys (3GPP TS 23.501 [2] and TS 23.502 [3]) and will not have any impact to the system architecture.

[0025] 2) Study architectural and functional implications on existing SA6 application enablers, e.g., ADAES and other SEAL services, EDGEAPP, for supporting AI / ML lifecycle operations, e.g., operations including the data collection, data preparation, training / inference / federated learning for the ML models to be used for ADAE layer analytics.

[0026] NOTE 2: For the UE data collection aspects, this study ..will rely upon EVEX framework specifications in 3GPP TS 26.531 v18.0.0 and 3GPP TS 26.532 v17.2.0.

[0027] 3) Identify potential solutions, including the information flows and developerfriendly application enablement APIs, where necessary to satisfy the architectural requirements and enhancements identified in bullets 1) and 2).

[0028] 4) Investigate possible impacts of application layer support for AI / ML services for different deployments and business models.

[0029] SUMMARY

[0030] As part of developing embodiments herein a problem was identified by the inventor and will first be discussed.

[0031] In SA2 Rel-18 AIMLsys, 5GS supports AI / ML-based services in application layer with the assumptions, for example, application AI / ML operation logic is controlled by an Application Function (AF), and AF request to the 5G System, e.g., NEF, NWDAF, etc., in the context of 5GS assistance to Application AI / ML operation. For application layer FL, there may one application server, e.g., an AF, controlling the application layer DML / FL / Split Learning training process among multiple application clients, e.g., UEs. If the application client acts as a coordinator for DML / FL / Split Learning training, it may coordinate the DML / FL / Split Learning among multiple the application servers.

[0032] In SA2 Rel-18 eNA_Ph3, FL among multiple NWDAFs within 5GC. The FL operations are performed among multiple NWDAFs containing Model Training Logical Function (MTLF). In detail, one FL Server NWDAF, e.g., NWDAF containing MTLF, controls the FL training process among multiple NWDAFs containing MTLF, e.g., FL Client NWDAFs.

[0033] The existing SEAL service ADAES, which corresponds to the NWDAF containing Analytics logical function (AnLF) defined in TS 23.288 v18.3.0, can provide analytics service. However, according to the system architecture capabilities developed by SA2 for FL among multiple NWDAFs in eNA_Ph3 (3GPP TS 23.288 v18.3.0, to enable the support of AI / ML service, the function corresponding to the NWDAF containing MTLF, which is the key functionality to organize, manage, and / or execute ML / DML / FL / Split Learning operations, is missing among the existing SEAL services, 3GPP TS 23.434 v18.6.0.ln the existing SEAL services, 3GPP TS 23.434 18.6.0, the training function corresponding to the NWDAF containing MTLF, which is the key part to organize, management, and / or execute ML / DML / FL / Split Learning operations, is missing.

[0034] An object of embodiments herein is to improve the performance of the communications network by providing a more efficient AI / ML training and management.

[0035] According to an aspect of embodiments herein, the object is achieved by a method performed by a managing node for training and / or managing a machine learning, ML, model associated with a vertical application layer, VAL, service in a communications network.

[0036] The managing node manages one or more operations in the communications network. The one or more operations are related to the ML model.

[0037] The managing node executes at least one out of the managed one or more operations in the communications network.

[0038] According to another aspect of embodiments herein, the object is achieved by a method performed by an executing node for training a machine learning, ML, model associated with a vertical application layer, VAL, service in a communications network.

[0039] The executing node executes one or more operations in the communications network. The one or more operations are related to training the ML model. According to another aspect of embodiments herein, the object is achieved by a method performed by a requesting node for training, managing and / or providing a machine learning, ML, model associated with a vertical application layer, VAL, service in a communications network.

[0040] The requesting node requests a managing node and / or an execution node to perform one or more operations in the communications network. The one or more operations are related to managing, training and / or providing the ML model.

[0041] According to another aspect of embodiments herein, the object is achieved by a managing node configured to train and / or manage a machine learning, ML, model associated with a vertical application layer, VAL, service in a communications network. The managing node is further configured to:

[0042] Manage one or more operations in the communications network. The one or more operations are adapted to be related to the ML model.

[0043] Execute at least one out of the managed one or more operations in the communications network.

[0044] According to another aspect of embodiments herein, the object is achieved by an executing node configured to train a machine learning, ML, model associated with a vertical application layer, VAL, service in a communications network. The executing node is further configured to:

[0045] Executing one or more operations in the communications network. The one or more operations are adapted to be related to training the ML model.

[0046] According to another aspect of embodiments herein, the object is achieved by a requesting node configured to train, manage and / or provide a machine learning, ML, model associated with a vertical application layer, VAL, service in a communications network. The requesting node is further configured to:

[0047] Request a managing node and / or an execution node to perform one or more operations in the communications network. The one or more operations are adapted to be related to managing, training and / or providing the ML model.

[0048] Thanks to that the managing node manages the one or more operations, it is possible to efficiently train and manage ML models in the communication network.

[0049] Embodiments herein bring the advantage of an efficient mechanism improving the performance in the communication network. This is achieved by a more efficient managing and / or coordination of ML models, where the managing node manages one or more operations related to the ML model. This leads to an efficient ML model training and management, and results in an improved performance of the communication network. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Examples of embodiments herein are described in more detail with reference to attached drawings in which:

[0051] Figure 1 is a schematic block diagram illustrating embodiments of a wireless communications network.

[0052] Figure 2 is a flowchart depicting embodiments of a method in a managing node.

[0053] Figure 3 is a flowchart depicting embodiments of a method in an executing node.

[0054] Figure 4 is a flowchart depicting embodiments of a method in a requesting node.

[0055] Figure 5 is a schematic block diagram illustrating examples of embodiments herein,

[0056] Figures 6a-c are schematic block diagrams illustrating examples of embodiments herein,

[0057] Figures 7a-b are schematic block diagrams illustrating examples of embodiments herein,

[0058] Figures 8a-b are schematic block diagrams illustrating examples of embodiments herein,

[0059] Figures 9a-b are schematic block diagrams illustrating examples of embodiments herein.

[0060] Figure 10a is a schematic signaling diagram illustrating examples of embodiments herein.

[0061] Figure 10b is a schematic signaling diagram illustrating examples of embodiments herein.

[0062] Figure 11 is a schematic block diagram illustrating examples of embodiments herein.

[0063] Figure 12 is a schematic block diagram illustrating embodiments of a managing node.

[0064] Figure 13 is a schematic block diagram illustrating embodiments of an execution node.

[0065] Figure 14 is a schematic block diagram illustrating embodiments of a requesting node.

[0066] Figure 15 shows an example of a communication system QQ100 in accordance with some embodiments.

[0067] Figure 16 shows a UE QQ200 in accordance with some embodiments.

[0068] Figure 17 shows a network node QQ300 in accordance with some embodiments.

[0069] Figure 18 is a block diagram of a host QQ400, which may be an embodiment of the host QQ116 of Fig. 18, in accordance with various aspects described herein.

[0070] Figure 19 is a block diagram illustrating a virtualization environment QQ500 in which functions implemented by some embodiments may be virtualized. Figure 20 shows a communication diagram of a host QQ602 communicating via a network node QQ604 with a UE QQ606 over a partially wireless connection in accordance with some embodiments.

[0071] DETAILED DESCRIPTION

[0072] Embodiments herein relate to a communications network and the training and / or managing of a machine learning model associated with a vertical application layer.

[0073] As mentioned above, an object of embodiments herein is to improve the performance of the communications network by providing a more efficient AI / ML training and management.

[0074] This may e.g., be achieved by introducing a new function for AI / ML training and management. The new function could be collocated with the exiting application layer architecture, e.g., ADAES / ADAEC, etc., or a standalone new SEAL service, or a new enabler layer.

[0075] A Machine Learning Model Training and Management Enablement (MTME) for organizing, managing, and / or executing the ML / DML / FL / Split Learning operations is introduced with the following considerations.

[0076] - Support the application specific / VAL layer in organizing, managing, and / or executing ML / DML / FL / Split Learning model transfer, distribution and / or training covering both on-network and off-network models.

[0077] - Support flexible deployment of the new MTME function. The MTME function may be collocated with the exiting SEAL layer architecture, e.g., ADAES / ADAEC or any other existing SEAL service(s), a new standalone SEAL service, or a new enabler layer.

[0078] - The support of the following use-cases (non-exclusive list), e.g.:

[0079] - Distributed the video / voice / object detection and recognition model training for the robot swarm, e.g., multiple robots in an area adapt the video / voice / object detection and recognition models to the specific area conditions, such as e.g., weather, day / night cycle, environment, etc., in the real-time, and / or

[0080] - ML model training for the adaptive geofencing boundaries for vertical applications, e.g., delivery robots’ route distribution.

[0081] The MTME function contains two main functionals: - Management / Coordination functional:

[0082] - Manage / coordinate the training of ML model(s) relevant operations, include data collection, data preparation, training / inference (ML / DML / FL / Split Learning), etc. For DML / FL / Split Learning, this functional enable:

[0083] - DML / FL / Split Learning group member discovery and selection.

[0084] - Exchange of initial, intermediate, and final ML model(s) / ML model parameters and training information among DML / FL / Split Learning group members.

[0085] - Monitoring the training process and model performance.

[0086] - Maintaining the training process and DML / FL / Split Learning group members when dynamic changes of status among the DML / FL / Split Learning group members.

[0087] - Manage / coordinate the distribution of ML model(s).

[0088] - Manage / coordinate the transfer of ML model(s) / training / inference.

[0089] - Manage / coordinate the interactions among application server(s), application client(s), 5GC NFs (e.g., NEF, NWDAF, etc.), OAM, and other Network Entities for the ML operations (e.g., data collection, data preparation, ML model training / inference / distribution, ML model(s) / training / inference transfer).

[0090] - Execution / Computation functional:

[0091] - Execute the training of ML model(s) relevant operations.

[0092] - Execute the distribution of ML model(s).

[0093] - Execute the transfer of ML model(s) / training / inference.

[0094] The MTME function may be collocated with the exiting SEAL layer architecture, e.g., ADAES / ADAEC or any other existing SEAL service(s), a new standalone SEAL service, or a new enabler layer.

[0095] Figure 1 is a schematic overview depicting a communications network 100 wherein embodiments herein may be implemented. The wireless communications network 100 comprises one or more RANs and one or more CNs. The wireless communications network 100 may use a number of different technologies, such as Wi-Fi, Long Term Evolution (LTE), LTE-Advanced, 5G, New Radio (NR), 6G, Wideband Code Division Multiple Access (WCDMA), Global System for Mobile communications / enhanced Data rate for GSM Evolution (GSM / EDGE), or Ultra Mobile Broadband (UMB), just to mention a few possible implementations. Embodiments herein relate to recent technology trends that are of particular interest in a 5G context, however, embodiments are also applicable in further development of the existing wireless communication systems such as e.g.

[0096] WCDMA and LTE.

[0097] A number of RAN nodes operate in the communications network 100 such as e.g., a base station 105. The base station 105 provides radio coverage in a number of cells which may also be referred to as a beam or a beam group of beams, such as a cell 11 and a cell 12.

[0098] The base station 105 may be any of an NG-RAN node, a transmission and reception point e.g. a base station, a radio access network node such as a Wireless Local Area Network (WLAN) access point or an Access Point Station (AP STA), an access controller, a base station, e.g. a radio base station such as a NodeB, an evolved Node B (eNB, eNode B), a gNB, a base transceiver station, a radio remote unit, an Access Point Base Station, a base station router, a transmission arrangement of a radio base station, a stand-alone access point or any other network unit capable of communicating with a UE, such as e.g., a UE 120, within the service area served by the base station 105 depending e.g. on the radio access technology and terminology used. The base station 105 may be referred to as a serving RAN node and communicates with UEs such as the UE 120, with Downlink (DL) transmissions to the UE 120, and in Uplink (UL) transmissions from the UE 120.

[0099] A number of network nodes, such as e.g., a managing node 110, an executing node 121 and a requesting node 130, operate in the communications network 100. The managing node 110 and the executing node 121 respectively comprises, or implements, MTME functions. The managing node 110, the executing node 121 and the requesting node 130 may be located in the CN and / or the RAN respectively. Alternatively, or additionally, any of the managing node 110, the executing node 121 and the requesting node 130 may be comprised, or implemented, in the e.g., the UE 120. The managing node 110 may e.g., implement, such as comprise, an MTME function. The MTME function may comprise one or more of managing and / or coordinating functions, and executing and / or computation functions as described herein. The executing node 121 may e.g., implement, such as comprise, an MTME function. The MTME function may comprise one or more of managing and / or coordinating functions, and executing and / or computation functions. The requesting node 130 may e.g., implement, such as comprise, a VAL server and / or a VAL client.

[0100] A number of UEs, such as e.g., the UE 120, operate in the communication network 100. The UE 120 may also be referred to as an loT device, a mobile station, a non-access point (non-AP), a STA, and / or a wireless terminal. It should be understood by the skilled in the art that “UE” is a non-limiting term which means any terminal, wireless communication terminal, user equipment, Machine Type Communication (MTC) device, Device to Device (D2D) terminal, a radio device in a vehicle, or node e.g., smart phone, laptop, mobile phone, sensor, relay, mobile tablets or even a small base station communicating within a cell.

[0101] Methods herein may be performed by the managing node 110, the executing node 121 and the requesting node 130.. As an alternative, a Distributed Node (DN) and functionality, e.g. comprised in the cloud 150 as shown in Figure 1 , may be used for performing or partly performing the methods herein.

[0102] According to examples of embodiments herein, a method performed by a managing node 110 for managing and / or training an ML model associated with a VAL service is provided. Further, an execution node 121 for training an ML model associated with a VAL service is provided. Yet further, a requesting node 130 for training, managing and / or providing an ML model associated with a VAL service is provided. The methods, managing node 110, execution node 121 and requesting node 130 may e.g., overcome the above mentioned problem.

[0103] A number of embodiments will now be described, some of which may be seen as alternatives, while some may be used in combination.

[0104] The embodiments of a method will be generally described in view of the managing node 110 together with Figure 2, in view of the executing node 121 together with Figure 3 and in view of the requesting node 130 in view of Figure 4. This will be followed by a more detailed description.

[0105] A method according to embodiments herein will now be described from the view of the managing node 110, together with Figure 2. Figure 2 depicts example embodiments of a method performed by the managing node 110 for training and / or managing an ML model associated with a VAL service in the communications network 100. The managing node 110 may be, such as e.g., being implemented as, any one out of a standalone SEAL function, collocated with an existing SEAL function, or a new enabler layer. The managing node 110 may e.g., implement, such as comprise, an MTME function, e.g., an MTME server and / or an MTME client. The MTME function may comprise one or more of managing and / or coordinating functions, and executing and / or computation functions. The method comprises the following actions, which actions may be taken in any suitable order. Actions that are optional are presented in dashed boxes in Figure 2.

[0106] Action 201

[0107] The managing node 110 manages one or more operations in the communications network 100. The one or more operations are related to the ML model, such as e.g., training the ML model. In other words, the managing node 100 manages one or more ML model related operations.

[0108] The one or more managed operations may comprise any one or more out of:

[0109] - Manage and / or coordinate the training of the ML models

[0110] - Manage and / or coordinate the distribution of the ML model.

[0111] - Manage and / or coordinate the transfer of the ML model, training and / or inference.

[0112] - Manage and / or coordinate the interactions among application servers such as executing nodes 121 , application clients such as executing nodes 121 , network functions such as e.g., 5GC NFs, OAM nodes, and other network entities for the ML operations.

[0113] - Manage and or coordinate incoming request for ML model training and / or management from the requesting node 130.

[0114] Thus, the managing node 110 may manage any one or more of the managed operations mentioned above.

[0115] In some embodiments, managing and / or coordinating the training of the ML model comprises any one or more out of:

[0116] - Discovering Distributed ML, DML, Federated Learning, FL, and / or Split Learning group members

[0117] - selecting DML, FL and / or Split Learning group members,

[0118] - exchanging ML models, ML model parameters and / or training data among selected DML, FL and / or Split Learning group members, wherein the ML models ML model parameters and / or training data comprises any one or more out of initial, intermediate and final ML models, ML model parameters and / or training data,

[0119] - monitoring the training process and model performance, and

[0120] - maintaining the training process and DML, FL and / or Split Learning group members when dynamic changes of status among the DML, FL and / or Split Learning group members occurs.

[0121] Thus, managing and / or coordinating the training of the ML models may e.g., comprise DML, FL and / or Split Learning group member discovery and selection, exchange of initial, intermediate, and final ML models and / or ML model parameters and training information among DML, FL and / or Split Learning group members, monitoring the training process and model performance, and / or maintaining the training process and DML, FL and / or Split Learning group members when dynamic changes of status among the DML, FL and / or Split Learning group members. The group members may e.g., comprise one or more executing node 121 and / or one or more other managing nodes 110 acting as executing nodes 121.

[0122] Action 202

[0123] The managing node 110 executes at least one out of the managed one or more operations in the communications network 100.

[0124] Executing at least one of the one or more managed operations may e.g., comprise any one or more out of:

[0125] - Execute the training of ML models relevant operations.

[0126] - Execute the distribution of ML model.

[0127] - Execute the transfer of ML model, training and / or inference.

[0128] This may e.g., mean that the managing node 110 executes any of the one or more managed operations above. E.g., the managing node 110 may receive a request for ML model training and / or management from the requesting node 130 and discover and / or select one or more DML, FL and / or Split Learning group members, such as one or more executing nodes 121, for the ML model training. Further, the managing node 110 may distribute an initial ML model to be trained by the selected group members, and receive interim, or intermediate, ML models and / or ML model data, trained by the selected group members. In some examples, the managing node 110 also performs, or executes, ML model training of the initial ML model. The managing node 110 may update the initial ML model based on the received interim ML models and / or ML model data, e.g., by aggregating the received interim ML models and / or data. In some examples, the managing node performs, or executes, a ML model performance evaluation of the interim ML model in order to determine whether satisfies a training criterion. The training criterion may e.g., comprise a performance criterion and / or a training termination condition, such as a time limit or max number of training iterations. If the interim ML model does not satisfy the training criteria, the managing node 110 may manage the training of the ML model by performing, or executing, a further iteration of the training, thus repeating the distributing, receiving, training, updating and evaluating of the ML model described above, but this time with the interim ML model instead of the initial ML model. With the proviso that the ML model satisfies the training criterion, the managing node 110 may notify the requesting node 130 that the training is completed, and also in some examples transfer the final, or global, ML model to the requesting node 130.

[0129] A method according to embodiments herein will now be described from the view of the executing node 121 , together with Figure 3. Figure 3 depicts example embodiments of a method performed by an executing node 121 for training an ML model associated with a VAL service in the communications network 100. The executing node 121 may be, such as e.g., being implemented as, any one out of a standalone SEAL function, collocated with an existing SEAL function, or a new enabler layer. The executing node 121 may e.g., implement, such as comprise, an MTME function, e.g., an MTME server and / or an MTME client. The MTME function may comprise one or more of managing and / or coordinating functions, and executing and / or computation functions. The method comprises the following actions, which actions may be taken in any suitable order. Actions that are optional are presented in dashed boxes in Figure 3.

[0130] Action 301

[0131] The managing node 110 executes one or more operations in the communications network 100. The one or more operations are related to training the ML model.

[0132] Executing one or more operations may e.g., comprise any one or more out of:

[0133] - Execute the training of ML models relevant operations.

[0134] - Execute the distribution of ML model.

[0135] - Execute the transfer of ML model, training and / or inference.

[0136] This may e.g., mean that the executing node 121 executes any of the one or more operations above. In some examples, the executing node 121 may receive an initial an initial ML model to be trained from the managing node 130. The executing node 121 may perform, or execute, ML model training operation to train the ML model, and send the trained interim ML model, or ML model data, to the managing node 130. In some examples, e.g., when the interim ML model created by the managing node 130 by updating the initial ML model based one or more trained interim ML models received by one or more executing nodes 121, does not satisfy the training criteria, the executing node 121 may receive the interim ML model created by the managing node 110 and perform, or execute, a further training iteration. When the further training iteration is performed, the executing node 121 sends the trained interim ML model to the managing node 110.

[0137] In other examples, no managing node 110 takes part in the training. The executing node 121 may then exchange, or distribute, the trained interim ML model with one or more other executing nodes 121 , and each executing node 121 may update the initial ML model, or interim ML model depending on whether it is the first training iteration or not, e.g., by aggregating the received interim ML models and / or data, the executing node 121 may then perform, or execute, a ML model performance evaluation of the interim ML model in order to determine whether satisfies a training criterion. The training criterion may e.g., comprise a performance criterion and / or a training termination condition, such as a time limit or max number of training iterations. If the interim ML model does not satisfy the training criteria, the managing node 110 may manage the training of the ML model by performing, or executing, a further iteration of the training, thus repeating the distributing, receiving, training, updating and evaluating of the ML model described above. With the proviso that the ML model satisfies the training criterion, the executing node 121 may notify the requesting node 130 that the training is completed, and also in some examples transfer the final, or global, ML model, ML model parameters and / or training data to the requesting node 130. In this example, the executing node may also receive the request from the requesting node 130.

[0138] A method according to embodiments herein will now be described from the view of the requesting node 130, together with Figure 4. Figure 4 depicts example embodiments of a method performed by the requesting node 130 for training, managing and / or providing the ML model associated with a VAL service in the communications network 100. The requesting node 130 may e.g., implement, such as comprise, a VAL server and / or a VAL client. The method comprises the following actions, which actions may be taken in any suitable order. Actions that are optional are presented in dashed boxes in Figure 3.

[0139] Action 401

[0140] The requesting node 130 requests the managing node 110 and / or the execution node 121 to perform one or more operations in the communications network. The one or more operations are related to managing, training and / or providing the ML model.

[0141] In some embodiments, the one or more operation may comprise any one or more out of training and / or providing the ML model, managing distribution of the ML model, and managing transfer for the ML model.

[0142] In some examples, the requesting node 130 requests the managing node 110 and / or the executing node 121 to train and / or manage the ML model. The request may e.g., comprise requirements on the training, such as training filter information and / or a training criterion. The training criterion may e.g., comprise a target ML model performance level and / or a training terminate condition, such as a training time limit and / or a max number of training iterations.

[0143] In some embodiments, the requesting may further comprise receiving a response from the managing node 110 and / or executing node 121. The response acknowledges the request.

[0144] Action 402

[0145] In some embodiments, the requesting node 130 may receive a notification from the managing node 110 and / or the executing node 121. The notification may notify the requesting node 130 that the requested training is completed. In some examples, the notification may comprise the trained ML model and / or ML model data or parameters.

[0146] Embodiments mentioned above will now be further described and exemplified. The embodiments below are applicable to and may be combined with any suitable embodiment described above.

[0147] Figure 5 shows consumer, such as the requesting node 130, requesting an application server / client, such as the managing node 110 and / or the executing node 121, for ML model training and / or management service.

[0148] As shown in Figure 5, the consumer requests application server / client for ML model training and management service. The consumer may be VAL server / client, a 5GC NF, an AF, and / or SEAL server / client. The consumer may request the application server / client to e.g., train and / or provide ML model(s), management of the distribution of ML model(s), and / or management of the transfer of ML model(s).

[0149] After receiving the request from the consumer, the application server / client decides whether to provide the required service to the consumer. The application server / client comprises or may have access to the MTME function. If decided to provide the required service, the application server / client may:

[0150] Provide the required ML model(s) to the consumer if available,

[0151] Train ML model(s) and provide the trained ML model(s) to the consumer. For the training of ML model(s), the application server / client may: o execute the training of the required ML model(s) itself, and / or o organize / coordinate multiple application server(s) / client(s) to train the required ML model(s) using, e.g., DML / FL / Split Learning techniques.

[0152] Manage the distribution of ML model(s) according to the requirement of the consumer.

[0153] Manage the transfer of ML mode(s) according to the requirement of the consumer.

[0154] Figures 6a-c shows DML / FL / Split Learning among application server(s) and client(s) (a) on-network model and with the coordinate of application server, (b) off- network model and centralized with the coordinate of one application client, and (c) off- network model and decentralized without coordinator.

[0155] Figures 6a-c illustrates DML / FL / Split Learning among multiple application clients, the application server(s) / client(s) are with or can access to the MTME function for training ML model(s).

[0156] In Figure 6a, an application server, such as the managing node 110, with the MTME function with both Management / Coordination function and Execution / Computation function organizes and / or coordinates a DML / FL / Split Learning training process among multiple application clients, such as multiple execution clients 121, which comprises the MTME function with Execution / Computation function, and in some examples also Management / Coordination function.

[0157] The application server trains a global model by aggregating local models partially trained by each application client based on the iterative model averaging. Within each training iteration, an application client performs the training based on the model provided by the application server using local training data. Then the application client(s) reports the interim training results, e.g., intermediate ML model(s) / ML model parameters / training information, to the application server. The application server aggregates the received ML model(s) / parameters / information from the application clients, and updates the global model. Next, the updated global model is distributed to the application clients, the application clients may perform the training for the next iteration. In the on-network model, the application server may interact with the 5GC NFs to request and receive assistance information for the application layer AL / ML operations.

[0158] In Figure 6b one of the application client, application client #X, such as the managing node 110, with more sufficient resources, e.g., computation, communication, and / or energy, and comprising the MTME function with Management / Coordination function and Execution / Computation function organizes / coordinates a DML / FL / Split Learning training process among multiple application clients comprising the MTME function with Execution / Computation function, and in some examples Management / Coordination function. The DML / FL / Split Learning training process between the application client #X and the multiple application clients is the same as that for Figure 6a, by replacing the application server with application client #X. In the off-network model, the DML / FL / Split Learning training process are completed among the application clients with direct device-to-device connections, and without the assistance information from 5GC network.

[0159] In Figure 6c there is no coordinator for the DML / FL / Split Learning training process. The participating application clients, such as the execution nodes 121, comprises the MTME function with Execution / Computation function, and in some examples Management / Coordination functional, and does not act as coordinator. In each iterative, each application client exchanges their intermediate training ML model(s) / parameter / information with the other application clients, and update its local interim ML model based on the received updates from the other application clients. In the off-network model, the DML / FL / Split Learning training process are completed among the application clients with direct device-to-device connections, and without the assistance information from 5GC network.

[0160] The MTME function, such as e.g., the managing node 110 and / or the executing node 121 , provides application layer AI / ML services, organization, management, and / or execution of the application layer ML / DML / FL / Split Learning operations. The MTME function contains two main functionals, i.e. , Management / Coordination functional and Execution / Computation functional.

[0161] - Management / Coordination functional, such as e.g., the managing node 110 and / or the executing node 121 , may comprise one or more out of:

[0162] - Manage / coordinate the training of ML model(s) relevant operations, include data collection, data preparation, training / inference (ML / DML / FL / Split Learning), etc. For DML / FL / Split Learning, this functional enable:

[0163] - DML / FL / Split Learning group member discovery and selection.

[0164] - Exchange of initial, intermediate, and final ML model(s) / ML model parameters and training information among DML / FL / Split Learning group members.

[0165] - Monitoring the training process and model performance.

[0166] - Maintaining the training process and DML / FL / Split Learning group members when dynamic changes of status among the DML / FL / Split Learning group members. - Manage / coordinate the distribution of ML model(s).

[0167] - Manage / coordinate the transfer of ML model(s) / training / inference.

[0168] - Manage / coordinate the interactions among application server(s), application client(s), 5GC NFs (e.g., NEF, NWDAF, etc.), OAM, and other Network Entities for the ML operations (e.g., data collection, data preparation, ML model training / inference / distribution, ML model(s) / training / inference transfer).

[0169] - Execution / Computation functional, such as e.g., the managing node 110 and / or the executing node 121 , may comprise one or more out of:

[0170] - Execute the training of ML model(s) relevant operations.

[0171] - Execute the distribution of ML model(s).

[0172] - Execute the transfer of ML model(s) / training / inference.

[0173] The MTME function may be collocated with the exiting SEAL layer architecture, e.g., ADAES / ADAEC or any other existing SEAL service(s), a new standalone SEAL service, or a new enabler layer.

[0174] The MTME function may support the application specific / VAL layer in the organize, management, and / or execute ML / DML / FL / Split Learning model transfer / distribution / training covering both on-network and off-network models.

[0175] The MTME function may further support flexible deployment of the new function. The new function could be collocated with the exiting application layer architecture, e.g., ADAES / ADAEC, etc., or a standalone new SEAL service, or a new enabler layer.

[0176] Figure 7a-b shows a generic Functional Model with MTME as new SEAL service as (a) an on-network model and (b) an off-network model.

[0177] Figure 7a illustrates the generic on-network functional model with MTME as a new SEAL service. In the VAL, the VAL client communicates with the VAL server over VAL-UU reference point. VAL-UU supports both unicast and multicast delivery modes. The SEAL functional entities with MTME function on the UE and the server are grouped into AI / ML SEAL client(s), such as the execution node 121 , and AI / ML SEAL server(s), such as the managing node 110, respectively. The AI / ML SEAL with MTME function consists of a common set of services, e.g., ML model provision, manage / coordinate / execute ML model training, ML model distribution, and ML model transfer, and reference points. The AI / ML SEAL offers its services to the VAL.

[0178] The AI / ML SEAL client(s) communicates with the AI / ML SEAL server(s) over the AI / ML SEAL-UU reference points. AI / ML SEAL-UU supports both unicast and multicast delivery modes. The AI / ML SEAL client(s) provides the service enabler layer support functions to the VAL client(s) over SEAL-C reference points. The VAL server(s) communicate with the AI / ML SEAL server(s) over the AI / ML SEAL-S reference points. The AI / ML SEAL server(s) may communicate with the underlying 3GPP network systems using the respective 3GPP interfaces specified by the 3GPP network system.

[0179] Figure 7b illustrates the generic off-network functional model with MTME as a new SEAL service. In the VAL, the VAL client of UE1 communicates with VAL client of UE2 over VAL-PC5 reference point. An AI / ML SEAL client, such as the managing node 110 or the execution node 121, of UE1 interacts with the corresponding AI / ML SEAL client, such as the managing client 110 or the execution client 121 , of UE2 over AI / ML SEAL-PC5 reference points. The UE1, if connected to the network via Uu reference point, can also act as a UE-to-network relay, to enable UE2 to access the VAL server(s) over the VAL- UU reference point.

[0180] Figure 8a-b shows a generic Functional Model with MTME collocated with ADAE as (a) an on-network model and (b) an off-network model.

[0181] Figure 8a illustrates the generic on-network functional model with MTME collocated with ADAE. The ADAE client with MTME, such as the executing node 121 , communicates with the ADAE server with MTME, such as the managing node 110, over the AI / ML ADAE-UU reference point. The ADAE client with MTME provides the support for ML model provision, manage / coordinate / execute ML model training, ML model distribution, and ML model transfer functions to the VAL client(s) over ADAE-C reference point. The VAL server(s) communicates with the ADAE server with MTME over the AI / ML ADAE-S reference point. The ADAE server with MTME, acting as an AF, may communicate with the 5G Core Network functions, e.g., NEF, NWDAF, UPF, and GAM, e.g., over ADAE- OAM interface.

[0182] Figure 8b illustrates the generic off-network functional model with MTME collocated with ADAE. In the vertical application layer, the VAL client of UE1 communicates with VAL client of UE2 over VAL-PC5 reference point. An ADAE client with MTME, such as the managing node 110 or the executing node 121 , of UE1 interacts with the corresponding ADAE client with MTME, such as the managing node 110 or the executing node 121, of UE2 over AI / ML ADAE-PC5 reference points. The UE1 , if connected to the network via Uu reference point, can also act as a UE-to-network relay, to enable UE2 to access the VAL server(s) over the VAL-UU reference point. Figure 9a-b shows a generic functional model with MTME as a new enabler layer as (a) an on-network model and (b) an off-network model.

[0183] Figure 9a illustrates the generic on-network functional model with MTME as a new enabler layer. In the vertical application layer, the VAL client communicates with the VAL server over VAL-Ull reference point. VAL-Ull supports both unicast and multicast delivery modes. The MTME functional entities on the UE and the server are grouped into MTME client(s), such as the executing node 121 , and MTME server(s), such as the managing node 110, respectively. The MTME consists of a common set of services, e.g., ML model provision, manage / coordinate / execute ML model training, ML model distribution, and ML model transfer) and reference points. The MTME offers its services to the VAL.

[0184] The MTME client(s) communicates with the MTME server(s) over the MTME-UU reference points. MTME-UU supports both unicast and multicast delivery modes. The MTME client(s) provides the service enabler layer support functions to the VAL client(s) over MTME-C reference points. The VAL server(s) communicate with the MTME server(s) over the MTME-S reference points. The MTME server(s) may communicate with the underlying 3GPP network systems using the respective 3GPP interfaces specified by the 3GPP network system.

[0185] Figure 9b illustrates the generic off-network functional model with MTME as a new enabler layer. In the vertical application layer, the VAL client of UE1 communicates with VAL client of UE2 over VAL-PC5 reference point. An MTME client, such as the managing node 110 or the executing node 121 , of UE1 interacts with the corresponding MTME client, such as the managing node 110 or the executing node 121, of UE2 over MTME- PC5 reference points. The UE1, if connected to the network via Uu reference point, can also act as a UE-to-network relay, to enable UE2 to access the VAL server(s) over the VAL-UU reference point.

[0186] Figures 10a-b illustrate a general procedure for the MTME function to provide ML management / execution services to consumer and an example procedure for the MTME function to coordinate DML / FL / Split Learning and provide requested ML model to the consumer.

[0187] Figure 10a illustrates the general procedure for the MTME function, e.g., as a new SEAL service, collocated with ADAE, or as new enabler layer, to provide ML management / execution services to consumers. The procedure is described by the following steps: Step 1. The consumer, such as the requesting node 130, of the MTME server (MTMES), such as the managing node 110, service sends a VAL ML training and / or management subscription request to the MTME server.

[0188] Step 2. The MTMES sends a subscription response as a positive or negative acknowledgement to the consumer of the ML training / management service.

[0189] Step 3. The MTMES maps the analytics event ID to a corresponding ML management and / or execution type and determines a list of one or more ML operations. Such mapping may be determined by the MTMES based on the management and / or execution event type, vertical type and / or ML operation type.

[0190] Step 4. The MTMES performs the ML operations, e.g., manage, coordinate and / or execute ML model training, ML model distribution, ML model / training transfer.

[0191] Step 4a. The MTMES may perform, such as execute or manage, the ML operations itself.

[0192] Step 4b. The MTMES may interact with one or more MTME clients (MTMEC), such as the execution node 121, other SEAL servers, and / or 5GC NFs, e.g., NEF, NWDAF, etc., for data collection and / or assistance as studied in SA2 Rel-18 AIMLsys, for the ML operations.

[0193] Step 5. The MTMES send notification to the consumer for the requested service.

[0194] Figure 10b illustrates an example procedure for the MTME function to coordinate DML / FL / Split Learning and provide requested ML model to the consumer, such as the requesting node 130. The MTME function may be a new SEAL service, collocated with ADAE, or a new enabler layer. The MTMEs, such as the managing node 110, coordinate DML / FL / Split Learning among multiple MTMECs, such as multiple execution nodes 121. The procedure is described by the following steps:

[0195] Step 1. The consumer of the MTMES service sends a VAL ML training subscription request to MTMES, the subscription request may include detail requirements on the training, e.g., training filter information, target ML model performance level, training terminate condition, in the request.

[0196] Step 2. The MTMES sends a subscription response as a positive or negative acknowledgement to the consumer of the ML training service.

[0197] Step 3. The MTMES maps the analytics event ID to corresponding ML management and / or execution type, e.g., coordinate DML / FL / Split Learning, and determines a list of one or more ML operations, e.g., DML / FL / Split Learning relevant operations. Such mapping may be determined by the MTMES based on the management and / or execution event type, vertical type and / or ML operation type.

[0198] Step 4. The MTMES interacts with MTMEC(s), other SEAL servers, and / or 5GC

[0199] NFs for DML / FL / Split Learning preparation, including DML / FL / Split Learning group member discovery and selection, data collection and preparation.

[0200] Step 5. The MTMES sends an ML Training subscription request to the selected

[0201] MTMEC(s) for DML / FL / Split Learning training, the subscription request may include an initial ML model in the request.

[0202] Step 6. The MTMEC(s) sends a subscription response as a positive or negative acknowledgement to the MTMES.

[0203] Step 7. The MTMEC(s) trains local ML model(s) based on the initial / updated global

[0204] ML model / local determined ML model by using local dataset(s). Before or during the training process, the MTMEC(s) may collect data from the VAL client(s), etc.

[0205] Step 8. The MTMEC(s) sends intermediate ML model(s) / parameters / training information to the MTMES.

[0206] Step 9. The MTMES aggregates the received local ML model(s) / parameters to update global ML model.

[0207] Step 10. The MTMES updates the subscription to the MTMEC(s) to distribute the updated global ML model.

[0208] Steps 7-10 may be repeated until a target ML model performance level of the global ML model is achieved at the MTMES, or a training terminate conditions is reached.

[0209] Step 11. The MTMES send notification to the consumer, the notification may include the trained global ML model / parameters if requested by the consumer.

[0210] 1. Introduction

[0211] This pCR proposes new Key Issue for FS_AIMLAPP.

[0212] 2. Reason for Change

[0213] The New SID for study on application layer support for AI / ML services (SP-231182) was approved in SA#101 meeting. According to the approved SID, the objective of the study is to enable support for AI / ML services at application enablement layer, which include

[0214] 1) Analyse Rel-18 and Rel-19 requirements in 3GPP TS 22.261 related to AI / ML model distribution, transfer, training and further identify key issues, develop corresponding architectural requirements at the application enablement layer and potential enhancements to the application layer architecture

[0215] NOTE 1: This objective will leverage existing SA2 5GC capabilities developed in Re-18 AIMLsys (TS 23.501 and TS 23.502) and will not have any impact to the system architecture. 2) Study architectural and functional implications on existing SA6 application enablers (e.g. ADAES and other SEAL services, EDGEAPP) for supporting AI / ML lifecycle operations (e.g. operations including the data collection, data preparation, training / inference / federated learning for the ML models to be used for ADAE layer analytics).

[0216] NOTE 2: For the UE data collection aspects, this study will rely upon EVEX framework specifications in TS 26.531 and TS 26.532.

[0217] 3) Identify potential solutions, including the information flows and developer-friendly application enablement APIs, where necessary to satisfy the architectural requirements and enhancements identified in bullets 1) and 2).

[0218] 4) Investigate possible impacts of application layer support for AI / ML services for different deployments and business models.

[0219] Therefore, a new Key Issue is proposed for the study of the objective 1) and 2) on architecture and related functions.

[0220] 3. Conclusions

[0221] <Conclusion part (optional)>

[0222] 4. Proposal

[0223] It is proposed to agree the following changes to 3GPP TR 23.700-82.

[0224] X.Y Key Issue #1 : Support of Architecture and Functions for Application Layer AI / ML Services

[0225] SA2 Rel-18 AIMLsys (in 3GPP TS 23.501 and TS 23.502) studied 5G System support for AI / ML-based services in application layer with the assumptions, for example:

[0226] ■ Application AI / ML operation logic is controlled by an Application Function (AF).

[0227] ■ AF request to the 5G System in the context of 5GS assistance to Application AI / ML operation.

[0228] ■ For application layer Federated Learning, there may one application server (e.g., an AF) controling the application layer DML / FL training process among multiple application clients (e.g., UEs).

[0229] SA1 Rel-18 and Rel-19 identified requirements for the support of AI / ML model distribution, transfer, training for various applications, e.g., video / speech recognition, robot control, automotive, in 3GPP TS.22.874, TS 22.876, and TS 22.261. Such use cases and requirements from SA1 have application layer impacts, e.g., require the support of correspoinding architecture and functions to provide assistance in application layer AI / ML operations on model distribution, transfer and training.

[0230] In Rel-19, the new SID study on application layer support for AI / ML services, list the objective of the study include:

[0231] ■ Analyse Rel-18 and Rel-19 requirements in 3GPP TS 22.261 related to AI / ML model distribution, transfer, training and further identify key issues, develop corresponding architectural requirements at the application enablement layer and potential enhancements to the application layer architecture. NOTE 1: This objective will leverage existing SA2 5GC capabilities developed in Re-18 AIMLsys (TS 23.501 and TS 23.502) and will not have any impact to the system architecture.

[0232] ■ Study architectural and functional implications on existing SA6 application enablers (e.g. ADAES and other SEAL services, EDGEAPP) for supporting AI / ML lifecycle operations (e.g., operations including the data collection, data preparation, training / inference / federated learning for the ML models to be used for ADAE layer analytics).

[0233] NOTE 2: For the UE data collection aspects, this study will rely upon EVEX framework specifications in TS 26.531 and TS 26.532.

[0234] Therefore, to support the AI / ML services in application layer, the following aspects need to be studied:

[0235] How to support the architecture and related functions for application layer AI / ML services.

[0236] How to support the provision of application layer AI / ML services.

[0237] - How the above architecture and related functions supporting the management / execution of AI / ML lifecycle operations:

[0238] - How to support the AI / ML services in the application layer related to data collection and preparation for ML model training / inference / Federated learning.

[0239] - How to support the AI / ML services in the application layer related to Distributed ML / Federated Learning. It includes but is not limited to the following aspects:

[0240] - How to support the AI / ML services in the application layer to enable Federated Learning considering UE conditions.

[0241] - How to enable participation (like grouping) of clients in the application layer based on UE conditions (like capability).

[0242] - How to enable privacy-preserving and secure mechanism exchange of model updates between client and server over the 3GPP system.

[0243] How to ensure QoS for AI / ML service over 3GPP system.

[0244] How to provide UE status information regarding AI / ML operations to VAL server.

[0245] - How to enable analytics (prediction or statistics on e.g., model update time, communication time, latencies) to the server at the application layer.

[0246] - How to enable offloading computation tasks (e.g., ML model training) in 5G System (e.g., application enabler).

[0247] - How to support the AI / ML services in the application layer related to model distribution over the 3GPP system.

[0248] How to leverage the existing 5GC capabilities and assistance for the application layer AI / ML services.

[0249] How to uitilize the trained ML model in analytics. 1. Introduction

[0250] This pCR proposes a new solution for Key lssue#X for FS_AIMLAPP.

[0251] 2. Reason for Change

[0252] The New SID for study on application layer support for AI / ML services (SP-231182) was approved in SA#101 meeting. According to the approved SID, the study on how to support the architecture and related functions for application layer AI / ML services is required.

[0253] In SA2 Rel-18 AIMLsys, 5GS supports AI / ML-based services in application layer with the assumptions, for example, application AI / ML operation logic is controlled by an Application Function (AF), and AF request to the 5G System (e.g., NEF, NWDAF, etc.) in the context of 5GS assistance to Application AI / ML operation. For application layer FL, there may one application server (e.g., an AF) controlling the application layer DML / FL training process among multiple application clients (e.g., UEs).

[0254] In SA2 Rel-18 eNA_Ph3, FL among multiple NWDAFs within 5GC. The FL operations are performed among multiple NWDAFs containing MTLF (Model Training logical function). In detail, one FL Server NWDAF (NWDAF containing MTLF) controls the FL training process among multiple NWDAFs containing MTLF (FL Client NWDAFs).

[0255] The existing SEAL service ADAES, which corresponds to the NWDAF containing AnLF (Analytics logical function, defined in TS 23.288), can provide analytics service. However, according to the system architecture capabilities developed by SA2 for FL among multiple NWDAFs in eNA_Ph3 (in TS 23.288), to enable the support of AI / ML service, the function corresponding to the NWDAF containing MTLF (Model Training logical function, defined in TS 23.288), which is the key functionality to organize, manage, and / or execute ML / DML / FL operations, is missing among the existing SEAL services TS 23.434.

[0256] This paper proposes a solution on the Functional Architecture to Support Application Layer AI / ML Services.

[0257] 3. Conclusions

[0258] <Conclusion part (optional)>

[0259] 4. Proposal

[0260] It is proposed to agree the following changes to 3GPP TR 23.700-82.

[0261] X0 References

[0262] The following documents contain provisions which, through reference in this text, constitute provisions of the present document.

[0263] - References are either specific (identified by date of publication, edition number, version number, etc.) or non-specific.

[0264] - For a specific reference, subsequent revisions do not apply.

[0265] - For a non-specific reference, the latest version applies. In the case of a reference to a 3 GPP document (including a GSM document), a non-specific reference implicitly refers to the latest version of that document in the same Release as the present document.

[0266] [1] 3GPP TS 23.434: "Service Enabler Architecture Layer for Verticals (SEAL);

[0267] Functional architecture and information flows".

[0268] [2] 3GPP TS 23.401: "General Packet Radio Service (GPRS) enhancements for Evolved Universal Terrestrial Radio Access Network (E-UTRAN) access". [3] 3GPP TS 23.501: "System Architecture for the 5G System; Stage 2".

[0269] X1.Y1 Solution #Z: Functional Architecture to Support Application Layer AI / ML Services

[0270] X1.Y1.1 General

[0271] The following clauses specify generic functional model with ML Model Training and Management Enablement (MTME) as new SEAL service for support application layer AI / ML services.

[0272] X1.Y1.2 Functional Architecture

[0273] X1.Y1.2.1 General

[0274] The functional architecture for the ML model training and management enablement is based on the generic functional model specified in clause 6.2 of 3GPP TS 23.434 [1], It is organized into functional entities to describe a functional architecture which addresses the support for ML model training and management enablement aspects for vertical applications.

[0275] X1. Y1.2.2 On-Network Functional A rchitecture

[0276] Figure 11 illustrates the generic on-network functional model with ML Model Training and Management Enablement (MTME) as new SEAL service. In the vertical application layer, the VAL client communicates with the VAL server over VAL-UU reference point. VAL-UU supports both unicast and multicast delivery modes. The SEAL functional entities with MTME function on the UE and the server are grouped into AI / ML SEAL client(s) and AI / ML SEAL server(s) respectively. The AI / ML SEAL with MTME function consists of a common set of services (e.g., ML model provision, ML client registration management, ML client member management, training status estimation, ML model training execution) and reference points. The AI / ML SEAL offers its services to the vertical application layer (VAL).

[0277] The AI / ML SEAL client(s) communicates with the AI / ML SEAL server(s) over the AI / ML SEAL-UU reference points. AI / ML SEAL-UU supports both unicast and multicast delivery modes. The AI / ML SEAL client(s) provides the service enabler layer support functions to the VAL client(s) over SEAL-C reference points. The VAL server(s) communicate with the AI / ML SEAL server(s) over the AI / ML SEAL-S reference points. The AI / ML SEAL server(s) may communicate with the underlying 3GPP network systems using the respective 3GPP interfaces specified by the 3GPP network system.

[0278] X1.Y1.3 Functional Entities Description

[0279] X1.Y1.3.1 General

[0280] The functional entities for ML model training and management enablement SEAL service are described in the following subclauses.

[0281] X1. Y1.3.2 AI / ML SEAL client

[0282] The AI / ML SEAL client functional entity acts as the application client supporting ML model training and may supporting management of ML model training. It interacts with the AI / ML SEAL server and may provide ML model provision, ML model training execution, ML client registration management, ML client member management, training status estimation services. X1. Y1.3.2 AI / ML SEAL server

[0283] The AI / ML SEAL server functional entity provides for management and execution of ML model training supported within the vertical application layer. The AI / ML SEAL server supports providing ML model provision, ML client registration management, ML client member management, training status estimation, and ML model training execution services to the vertical application server and client(s).X1.Y1.4 Reference Points Description

[0284] X1.Y1.4.1 General

[0285] The reference points for the functional model for management and execution of ML model training are described in the following subclauses.

[0286] X1.Y1.4.2 AI / ML SEAL-UU

[0287] The interactions related to ML model training and management functions between the AI / ML SEAL client and AI / ML SEAL server are supported by AI / ML SEAL-UU reference point. This reference point utilizes Uu reference point as described in 3GPP TS 23.401 [2] and 3GPP TS 23.501 [3],

[0288] AI / ML SEAL-UU reference point is used for VAL service signalling for VAL service data management of the VAL service. The AI / ML SEAL-UU reference point supports:

[0289] - Configuration of ML model training and management related data at the AI / ML SEAL client by the AI / ML SEAL server; and

[0290] - Configuration of ML model training and management related data at the AI / ML SEAL server by the AI / ML SEAL client.

[0291] The AI / ML SEAL-UU reference point uses the HTTP-1 / HTTP-2 reference points for transport and routing of ML model training and management related signalling. The AI / ML SEAL-UU reference point uses the SI P-1 / SI P-2 reference points for subscription / notification related signalling.

[0292] X1.Y1.4.3 AI / ML SEALS

[0293] The interactions related to ML model training and management functions between the VAL server(s) and the AI / ML SEAL server are supported by AI / ML SEAL-S reference point.

[0294] AI / ML SEALS reference point supports the VAL server to obtain ML model / ML model training and management information corresponding to the VAL service. The AI / ML SEALS reference point uses HTTP-1 / HTTP-2 reference points fortransport and routing of ML model / ML model training and management related signalling. The AI / ML SEALS reference point uses SIPS reference point for subscription / notification related sig nailing. X1. Y1.4.4 SEAL-C

[0295] The interactions related to ML model training and management functions between the VAL client(s) and the AI / ML SEAI client within a VAL UE are supported by SEAL-C reference point.

[0296] Figure 12 shows an example of arrangement in the managing node 110.

[0297] The managing node 110 may comprise an input and output interface 1200 configured to communicate with each other. The input and output interface 1200 may comprise a receiver, e.g. wired and / or wireless, (not shown) and a transmitter, e.g. wired and / or wireless, (not shown). The embodiments herein may be implemented through a respective processor or one or more processors, such as at least one processor 1210 of a processing circuitry in the managing node 110 depicted in Figure 12, together with computer program code for performing the functions and actions of the embodiments herein. The program code mentioned above may also be provided as a computer program product, for instance in the form of a data carrier carrying computer program code for performing the embodiments herein when being loaded into the managing node 110. One such carrier may be in the form of a CD ROM disc. It is however feasible with other data carriers such as a memory stick. The computer program code may furthermore be provided as pure program code on a server and downloaded to the managing node 110.

[0298] The managing node 110 and / or the processor 1210 is e.g., configured to train and / or manage an ML model associated with a VAL service in the communications network 100. The managing node 121 and / or the processor 1210 is further configured to any one or more out of:

[0299] Manage one or more operations in the communications network, wherein the one or more operations are adapted to be related to the ML model.

[0300] Execute at least one out of the managed one or more operations in the communications network.

[0301] In some embodiments, the managing node 110 and / or processor 1210 is configured to manage the one or more or operation by further being figured to any one or more out of:

[0302] - Manage and / or coordinate the training of the ML model.

[0303] - Manage and / or coordinate the distribution of the ML model.

[0304] - Manage and / or coordinate the transfer of the ML model, training and / or inference.

[0305] - Manage and / or coordinate the interactions among one or more out of application servers, application clients, network functions, operation and maintenance nodes, and / or other network entities for the ML operations.

[0306] - Manage and / or coordinate incoming requests for ML model training and / or ML management from a requesting node (130).

[0307] In some embodiments, manage and / or coordinate the training of the ML model may be adapted to comprise any one or more out of:

[0308] - Discover DML, FL, and / or Split Learning group members.

[0309] - Select DML, FL and / or Split Learning group members.

[0310] - Exchange ML models, ML model parameters and / or training data among selected DML, FL and / or Split Learning group members. The ML models, ML model parameters and / or training data may be adapted to comprise any one or more out of initial, intermediate and final ML models, ML model parameters and / or training data.

[0311] - Monitor the training process and model performance.

[0312] - Maintain the training process and DML, FL and / or Split Learning group members when dynamic changes of status among the DML, FL and / or Split Learning group members occurs

[0313] In some embodiments, the managing node 110 and / or processor 1210 may be configured to execute the at least one managed operation by further being configured to any one or more out of:

[0314] - Execute the training of ML model relevant operations

[0315] - Execute the distribution of ML model.

[0316] - Execute the transfer of ML model, training and / or inference.

[0317] In some embodiments, the managing node 110 is adapted to be any one out of:

[0318] - A standalone SEAL function.

[0319] - Collocated with an existing SEAL function, or

[0320] - A new enabler layer.

[0321] The managing node 110 may further comprise respective a memory 1220 comprising one or more memory units. The memory 1220 comprises instructions executable by the processor 1210 in the managing node 110.

[0322] The memory 1220 is arranged to be used to store instructions, data, configurations, ML model, training data, and applications to perform the methods herein when being executed in the managing node 110.

[0323] In some embodiments, a computer program 1230 comprises instructions, which when executed by the at least one processor 1210, cause the at least one processor 1210 of the managing node 110 to perform the actions above.

[0324] In some embodiments, a respective carrier 1240 comprises the respective computer program 1230, wherein the carrier 1240 is one of an electronic signal, an optical signal, an electromagnetic signal, a magnetic signal, an electric signal, a radio signal, a microwave signal, or a computer-readable storage medium.

[0325] Thus, embodiments herein may disclose the managing node 110 configured to train and / or manage an ML model associated with a VAL service in the communications network 100. The managing node 110 comprises the processor 1210 and the memory 1220, said memory 1220 comprising instructions executable by said processor 1210 whereby said managing node 110 is operative to perform any of the methods herein. Those skilled in the art will also appreciate that the functional modules in the managing node 110, described below may refer to a combination of analog and digital circuits, and / or one or more processors configured with software and / or firmware, e.g. stored in the managing node 110, that when executed by the respective one or more processors such as the at least one processor 1210 described above cause the respective at least one processor 1210 to perform actions according to any of the actions above. One or more of these processors, as well as the other digital hardware, may be included in a single Application-Specific Integrated Circuitry (ASIC), or several processors and various digital hardware may be distributed among several separate components, whether individually packaged or assembled into a system-on-a-chip (SoC).

[0326] Figure 13 shows an example of arrangement in the executing node 121.

[0327] The executing node 121 may comprise an input and output interface 1300 configured to communicate with each other. The input and output interface 1300 may comprise a receiver, e.g. wired and / or wireless, (not shown) and a transmitter, e.g. wired and / or wireless, (not shown).

[0328] The embodiments herein may be implemented through a respective processor or one or more processors, such as at least one processor 1310 of a processing circuitry in the executing node 121 depicted in Figure 13, together with computer program code for performing the functions and actions of the embodiments herein. The program code mentioned above may also be provided as a computer program product, for instance in the form of a data carrier carrying computer program code for performing the embodiments herein when being loaded into the executing node 121. One such carrier may be in the form of a CD ROM disc. It is however feasible with other data carriers such as a memory stick. The computer program code may furthermore be provided as pure program code on a server and downloaded to the executing node 121.

[0329] The executing node 121 and / or processor 1310 is e.g., configured to train a ML model associated with a VAL service in the communications network 100. The executing node 121 is further configured to:

[0330] Executing one or more operations in the communications network, wherein the one or more operations are adapted to be related to training the ML model.

[0331] In some embodiments, the executing node 121 and / or processor 1310 may be configured to execute the at least one managed operation by further being configured to any one or more out of:

[0332] - Execute the training of ML model relevant operations. - Execute the distribution of ML model.

[0333] - Execute the transfer of ML model, training and / or inference.

[0334] In some embodiments, the executing node 121 may be adapted to be any one out of:

[0335] - A standalone SEAL function.

[0336] - Collocated with an existing SEAL function.

[0337] - A new enabler layer.

[0338] The executing node 121 may further comprise respective a memory 1320 comprising one or more memory units. The memory 1320 comprises instructions executable by the processor 1310 in the executing node 121.

[0339] The memory 1320 is arranged to be used to store instructions, data, configurations, ML model, training data, and applications to perform the methods herein when being executed in the executing node 121.

[0340] In some embodiments, a computer program 1330 comprises instructions, which when executed by the at least one processor 1310, cause the at least one processor 1310 of the executing node 121 to perform the actions above.

[0341] In some embodiments, a respective carrier 1340 comprises the respective computer program 1330, wherein the carrier 1340 is one of an electronic signal, an optical signal, an electromagnetic signal, a magnetic signal, an electric signal, a radio signal, a microwave signal, or a computer-readable storage medium.

[0342] Thus, embodiments herein may disclose the executing node 121 configured to train a ML model associated with a VAL service in the communications network 100 The executing node 121 comprises the processor 1310 and the memory 1320, said memory 1320 comprising instructions executable by said processor 1310 whereby said executing node 121 is operative to perform any of the methods herein.

[0343] Those skilled in the art will also appreciate that the functional modules in the executing node 121 , described below may refer to a combination of analog and digital circuits, and / or one or more processors configured with software and / or firmware, e.g. stored in the executing node 121 , that when executed by the respective one or more processors such as the at least one processor 1310 described above cause the respective at least one processor 1310 to perform actions according to any of the actions above. One or more of these processors, as well as the other digital hardware, may be included in a single Application-Specific Integrated Circuitry (ASIC), or several processors and various digital hardware may be distributed among several separate components, whether individually packaged or assembled into a system-on-a-chip (SoC). Figure 14 shows an example of arrangement in the requesting node 130.

[0344] The requesting node 130 may comprise an input and output interface 1400 configured to communicate with each other. The input and output interface 1400 may comprise a receiver, e.g. wired and / or wireless, (not shown) and a transmitter, e.g. wired and / or wireless, (not shown).

[0345] The embodiments herein may be implemented through a respective processor or one or more processors, such as at least one processor 1410 of a processing circuitry in the requesting node 130 depicted in Figure 14, together with computer program code for performing the functions and actions of the embodiments herein. The program code mentioned above may also be provided as a computer program product, for instance in the form of a data carrier carrying computer program code for performing the embodiments herein when being loaded into the requesting node 130. One such carrier may be in the form of a CD ROM disc. It is however feasible with other data carriers such as a memory stick. The computer program code may furthermore be provided as pure program code on a server and downloaded to the requesting node 130.

[0346] The requesting node 130 and / or processor 1410 is configured to train, manage and / or provide a ML model associated with a VAL service in the communications network 100. The requesting node 130 is further configured to:

[0347] Request a managing node 110 and / or an executing node 121 to perform one or more operations in the communications network. The one or more operations are adapted to be related to managing, training and / or providing the ML model.

[0348] In some embodiments, the one or more operation may be adapted to comprise any one or more out of:

[0349] - Train and / or providing the ML model.

[0350] - Manage distribution of the ML model.

[0351] - Manage transfer for the ML model.

[0352] In some embodiments, the requesting node 130 and / or processor 1410 may be further configured to:

[0353] Receive a notification from the managing node 110 and / or the executing node 121. The notification may be adapted to notify the requesting node 130 that the training is completed, and / or comprising the trained ML model.

[0354] In some embodiments, the requesting node 130 and / or processor 1410 may be configured to request the managing node 110 and / or the executing node 121 to perform one or more operations by further being configured to receive a response from the managing node 110 and / or executing node 121. The response may acknowledge the request.

[0355] The requesting node 130 may further comprise respective a memory 1420 comprising one or more memory units. The memory 1420 comprises instructions executable by the processor 1410 in the requesting node 130.

[0356] The memory 1420 is arranged to be used to store instructions, data, configurations, ML model, training data, and applications to perform the methods herein when being executed in the requesting node 130.

[0357] In some embodiments, a computer program 1430 comprises instructions, which when executed by the at least one processor 1430, cause the at least one processor 1410 of the requesting node 130 to perform the actions above.

[0358] In some embodiments, a respective carrier 1440 comprises the respective computer program 1430, wherein the carrier 1440 is one of an electronic signal, an optical signal, an electromagnetic signal, a magnetic signal, an electric signal, a radio signal, a microwave signal, or a computer-readable storage medium.

[0359] Thus, embodiments herein may disclose the requesting node 130 configured to train, manage and / or provide a ML model associated with a VAL service in the communications network 100. The requesting node 130 comprises the processor 1410 and the memory 1420, said memory 1420 comprising instructions executable by said processor 1410 whereby said requesting node 130 is operative to perform any of the methods herein.

[0360] Those skilled in the art will also appreciate that the functional modules in the requesting node 130, described below may refer to a combination of analog and digital circuits, and / or one or more processors configured with software and / or firmware, e.g. stored in the requesting node 130, that when executed by the respective one or more processors such as the at least one processor 1410 described above cause the respective at least one processor 1410 to perform actions according to any of the actions above. One or more of these processors, as well as the other digital hardware, may be included in a single Application-Specific Integrated Circuitry (ASIC), or several processors and various digital hardware may be distributed among several separate components, whether individually packaged or assembled into a system-on-a-chip (SoC).

[0361] ADDITIONAL EXPLANATION Some of the embodiments contemplated herein will now be described more fully with reference to the accompanying drawings. Embodiments are provided by way of example to convey the scope of the subject matter to those skilled in the art.

[0362] Figure 15 shows an example of a communication system QQ100 in accordance with some embodiments.

[0363] In the example, the communication system QQ100 includes a telecommunication network QQ102 that includes an access network QQ104, such as a radio access network (RAN), and a core network QQ106, which includes one or more core network nodes QQ108 (being examples of the network node 110). The access network QQ104 includes one or more access network nodes, such as network nodes QQ110a and QQ110b (one or more of which may be generally referred to as network nodes QQ110 being examples of the network node 110), or any other similar 3rd Generation Partnership Project (3GPP) access nodes or non-3GPP access points. Moreover, as will be appreciated by those of skill in the art, a network node is not necessarily limited to an implementation in which a radio portion and a baseband portion are supplied and integrated by a single vendor. Thus, it will be understood that network nodes include disaggregated implementations or portions thereof. For example, in some embodiments, the telecommunication network QQ102 includes one or more Open-RAN (ORAN) network nodes. An ORAN network node is a node in the telecommunication network QQ102 that supports an ORAN specification (e.g., a specification published by the O-RAN Alliance, or any similar organization) and may operate alone or together with other nodes to implement one or more functionalities of any node in the telecommunication network QQ102, including one or more network nodes QQ110 and / or core network nodes QQ108.

[0364] Examples of an ORAN network node include an open radio unit (0-Rll), an open distributed unit (0-Dll), an open central unit (O-CU), including an O-CU control plane (O- CLI-CP) or an O-CU user plane (O-CU-UP), a RAN intelligent controller (near-real time or non-real time) hosting software or software plug-ins, such as a near-real time control application (e.g., xApp) or a non-real time control application (e.g., rApp), or any combination thereof (the adjective “open” designating support of an ORAN specification). The network node may support a specification by, for example, supporting an interface defined by the ORAN specification, such as an A1 , F1 , W1, E1 , E2, X2, Xn interface, an open fronthaul user plane interface, or an open fronthaul management plane interface. Moreover, an ORAN access node may be a logical node in a physical node. Furthermore, an ORAN network node may be implemented in a virtualization environment (described further below) in which one or more network functions are virtualized. For example, the virtualization environment may include an O-Cloud computing platform orchestrated by a Service Management and Orchestration Framework via an 0-2 interface defined by the O-RAN Alliance or comparable technologies. The network nodes QQ110 facilitate direct or indirect connection of user equipment (UE), such as by connecting UEs QQ112a, QQ112b, QQ112c, and QQ112d (one or more of which may be generally referred to as UEs QQ112 being examples of the wireless device 121) to the core network QQ106 over one or more wireless connections.

[0365] Example wireless communications over a wireless connection include transmitting and / or receiving wireless signals using electromagnetic waves, radio waves, infrared waves, and / or other types of signals suitable for conveying information without the use of wires, cables, or other material conductors. Moreover, in different embodiments, the communication system QQ100 may include any number of wired or wireless networks, network nodes, UEs, and / or any other components or systems that may facilitate or participate in the communication of data and / or signals whether via wired or wireless connections. The communication system QQ100 may include and / or interface with any type of communication, telecommunication, data, cellular, radio network, and / or other similar type of system.

[0366] The UEs QQ112 may be any of a wide variety of communication devices, including wireless devices arranged, configured, and / or operable to communicate wirelessly with the network nodes QQ110 and other communication devices. Similarly, the network nodes QQ110 are arranged, capable, configured, and / or operable to communicate directly or indirectly with the UEs QQ112 and / or with other network nodes or equipment in the telecommunication network QQ102 to enable and / or provide network access, such as wireless network access, and / or to perform other functions, such as administration in the telecommunication network QQ102.

[0367] In the depicted example, the core network QQ106 connects the network nodes QQ110 to one or more hosts, such as host QQ116. These connections may be direct or indirect via one or more intermediary networks or devices. In other examples, network nodes may be directly coupled to hosts. The core network QQ106 includes one more core network nodes (e.g., core network node QQ108) that are structured with hardware and software components. Features of these components may be substantially similar to those described with respect to the UEs, network nodes, and / or hosts, such that the descriptions thereof are generally applicable to the corresponding components of the core network node QQ108. Example core network nodes include functions of one or more of a Mobile Switching Center (MSC), Mobility Management Entity (MME), Home Subscriber Server (HSS), Access and Mobility Management Function (AMF), Session Management Function (SMF), Authentication Server Function (ALISF), Subscription Identifier Deconcealing function (SIDF), Unified Data Management (UDM), Security Edge Protection Proxy (SEPP), Network Exposure Function (NEF), and / or a User Plane Function (UPF).

[0368] The host QQ116 may be under the ownership or control of a service provider other than an operator or provider of the access network QQ104 and / or the telecommunication network QQ102, and may be operated by the service provider or on behalf of the service provider. The host QQ116 may host a variety of applications to provide one or more service. Examples of such applications include live and pre-recorded audio / video content, data collection services such as retrieving and compiling data on various ambient conditions detected by a plurality of UEs, analytics functionality, social media, functions for controlling or otherwise interacting with remote devices, functions for an alarm and surveillance center, or any other such function performed by a server.

[0369] As a whole, the communication system QQ100 of Figure 15 enables connectivity between the UEs, network nodes, and hosts. In that sense, the communication system may be configured to operate according to predefined rules or procedures, such as specific standards that include, but are not limited to: Global System for Mobile Communications (GSM); Universal Mobile Telecommunications System (UMTS); Long Term Evolution (LTE), and / or other suitable 2G, 3G, 4G, 5G standards, or any applicable future generation standard (e.g., 6G); wireless local area network (WLAN) standards, such as the Institute of Electrical and Electronics Engineers (IEEE) 802.11 standards (WiFi); and / or any other appropriate wireless communication standard, such as the Worldwide Interoperability for Microwave Access (WiMax), Bluetooth, Z-Wave, Near Field Communication (NFC) ZigBee, LiFi, and / or any low-power wide-area network (LPWAN) standards such as LoRa and Sigfox.

[0370] In some examples, the telecommunication network QQ102 is a cellular network that implements 3GPP standardized features. Accordingly, the telecommunications network QQ102 may support network slicing to provide different logical networks to different devices that are connected to the telecommunication network QQ102. For example, the telecommunications network QQ102 may provide Ultra Reliable Low Latency Communication (URLLC) services to some UEs, while providing Enhanced Mobile Broadband (eMBB) services to other UEs, and / or Massive Machine Type Communication (mMTC) / Massive loT services to yet further UEs.

[0371] In some examples, the UEs QQ112 are configured to transmit and / or receive information without direct human interaction. For instance, a UE may be designed to transmit information to the access network QQ104 on a predetermined schedule, when triggered by an internal or external event, or in response to requests from the access network QQ104. Additionally, a UE may be configured for operating in single- or multi- RAT or multi-standard mode. For example, a UE may operate with any one or combination of Wi-Fi, NR (New Radio) and LTE, i.e. being configured for multi-radio dual connectivity (MR-DC), such as E-UTRAN (Evolved-UMTS Terrestrial Radio Access Network) New Radio - Dual Connectivity (EN-DC).

[0372] In the example, the hub QQ114 communicates with the access network QQ104 to facilitate indirect communication between one or more UEs (e.g., UE QQ112c and / or QQ112d) and network nodes (e.g., network node QQ110b). In some examples, the hub QQ114 may be a controller, router, content source and analytics, or any of the other communication devices described herein regarding UEs. For example, the hub QQ114 may be a broadband router enabling access to the core network QQ106 for the UEs. As another example, the hub QQ114 may be a controller that sends commands or instructions to one or more actuators in the UEs. Commands or instructions may be received from the UEs, network nodes QQ110, or by executable code, script, process, or other instructions in the hub QQ114. As another example, the hub QQ114 may be a data collector that acts as temporary storage for UE data and, in some embodiments, may perform analysis or other processing of the data. As another example, the hub QQ114 may be a content source. For example, for a UE that is a VR headset, display, loudspeaker or other media delivery device, the hub QQ114 may retrieve VR assets, video, audio, or other media or data related to sensory information via a network node, which the hub QQ114 then provides to the UE either directly, after performing local processing, and / or after adding additional local content. In still another example, the hub QQ114 acts as a proxy server or orchestrator for the UEs, in particular if one or more of the UEs are low energy loT devices.

[0373] The hub QQ114 may have a constant / persistent or intermittent connection to the network node QQ110b. The hub QQ114 may also allow for a different communication scheme and / or schedule between the hub QQ114 and UEs (e.g., UE QQ112c and / or QQ112d), and between the hub QQ114 and the core network QQ106. In other examples, the hub QQ114 is connected to the core network QQ106 and / or one or more UEs via a wired connection. Moreover, the hub QQ114 may be configured to connect to an M2M service provider over the access network QQ104 and / or to another UE over a direct connection. In some scenarios, UEs may establish a wireless connection with the network nodes QQ110 while still connected via the hub QQ114 via a wired or wireless connection. In some embodiments, the hub QQ114 may be a dedicated hub - that is, a hub whose primary function is to route communications to / from the UEs from / to the network node QQ110b. In other embodiments, the hub QQ114 may be a non-dedicated hub - that is, a device which is capable of operating to route communications between the UEs and network node QQ110b, but which is additionally capable of operating as a communication start and / or end point for certain data channels.

[0374] Figure 16 shows a UE QQ200 in accordance with some embodiments. As used herein, a UE refers to a device capable, configured, arranged and / or operable to communicate wirelessly with network nodes and / or other UEs. Examples of a UE include, but are not limited to, a smart phone, mobile phone, cell phone, voice over IP (VoIP) phone, wireless local loop phone, desktop computer, personal digital assistant (PDA), wireless cameras, gaming console or device, music storage device, playback appliance, wearable terminal device, wireless endpoint, mobile station, tablet, laptop, laptop- embedded equipment (LEE), laptop-mounted equipment (LME), smart device, wireless customer-premise equipment (CPE), vehicle, vehicle-mounted or vehicle embedded / integrated wireless device, etc. Other examples include any UE identified by the 3rd Generation Partnership Project (3GPP), including a narrow band internet of things (NB-loT) UE, a machine type communication (MTC) UE, and / or an enhanced MTC (eMTC) UE.

[0375] A UE may support device-to-device (D2D) communication, for example by implementing a 3GPP standard for sidelink communication, Dedicated Short-Range Communication (DSRC), vehicle-to-vehicle (V2V), vehicle-to-infrastructure (V2I), or vehicle-to-everything (V2X). In other examples, a UE may not necessarily have a user in the sense of a human user who owns and / or operates the relevant device. Instead, a UE may represent a device that is intended for sale to, or operation by, a human user but which may not, or which may not initially, be associated with a specific human user (e.g., a smart sprinkler controller). Alternatively, a UE may represent a device that is not intended for sale to, or operation by, an end user but which may be associated with or operated for the benefit of a user (e.g., a smart power meter).

[0376] The UE QQ200 includes processing circuitry QQ202 that is operatively coupled via a bus QQ204 to an input / output interface QQ206, a power source QQ208, a memory QQ210, a communication interface QQ212, and / or any other component, or any combination thereof. Certain UEs may utilize all or a subset of the components shown in Figure QQ2. The level of integration between the components may vary from one UE to another UE. Further, certain UEs may contain multiple instances of a component, such as multiple processors, memories, transceivers, transmitters, receivers, etc.

[0377] The processing circuitry QQ202 is configured to process instructions and data and may be configured to implement any sequential state machine operative to execute instructions stored as machine-readable computer programs in the memory QQ210. The processing circuitry QQ202 may be implemented as one or more hardware-implemented state machines (e.g., in discrete logic, field-programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), etc.); programmable logic together with appropriate firmware; one or more stored computer programs, general-purpose processors, such as a microprocessor or digital signal processor (DSP), together with appropriate software; or any combination of the above. For example, the processing circuitry QQ202 may include multiple central processing units (CPUs).

[0378] In the example, the input / output interface QQ206 may be configured to provide an interface or interfaces to an input device, output device, or one or more input and / or output devices. Examples of an output device include a speaker, a sound card, a video card, a display, a monitor, a printer, an actuator, an emitter, a smartcard, another output device, or any combination thereof. An input device may allow a user to capture information into the UE QQ200. Examples of an input device include a touch-sensitive or presence-sensitive display, a camera (e.g., a digital camera, a digital video camera, a web camera, etc.), a microphone, a sensor, a mouse, a trackball, a directional pad, a trackpad, a scroll wheel, a smartcard, and the like. The presence-sensitive display may include a capacitive or resistive touch sensor to sense input from a user. A sensor may be, for instance, an accelerometer, a gyroscope, a tilt sensor, a force sensor, a magnetometer, an optical sensor, a proximity sensor, a biometric sensor, etc., or any combination thereof. An output device may use the same type of interface port as an input device. For example, a Universal Serial Bus (USB) port may be used to provide an input device and an output device.

[0379] In some embodiments, the power source QQ208 is structured as a battery or battery pack. Other types of power sources, such as an external power source (e.g., an electricity outlet), photovoltaic device, or power cell, may be used. The power source QQ208 may further include power circuitry for delivering power from the power source QQ208 itself, and / or an external power source, to the various parts of the UE QQ200 via input circuitry or an interface such as an electrical power cable. Delivering power may be, for example, for charging of the power source QQ208. Power circuitry may perform any formatting, converting, or other modification to the power from the power source QQ208 to make the power suitable for the respective components of the UE QQ200 to which power is supplied.

[0380] The memory QQ210 may be or be configured to include memory such as random access memory (RAM), read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), magnetic disks, optical disks, hard disks, removable cartridges, flash drives, and so forth. In one example, the memory QQ210 includes one or more application programs QQ214, such as an operating system, web browser application, a widget, gadget engine, or other application, and corresponding data QQ216. The memory QQ210 may store, for use by the UE QQ200, any of a variety of various operating systems or combinations of operating systems.

[0381] The memory QQ210 may be configured to include a number of physical drive units, such as redundant array of independent disks (RAID), flash memory, USB flash drive, external hard disk drive, thumb drive, pen drive, key drive, high-density digital versatile disc (HD-DVD) optical disc drive, internal hard disk drive, Blu-Ray optical disc drive, holographic digital data storage (HDDS) optical disc drive, external mini-dual in-line memory module (DIMM), synchronous dynamic random access memory (SDRAM), external micro-DIMM SDRAM, smartcard memory such as tamper resistant module in the form of a universal integrated circuit card (UICC) including one or more subscriber identity modules (SIMs), such as a USIM and / or ISIM, other memory, or any combination thereof. The UICC may for example be an embedded UICC (eUlCC), integrated UICC (iUICC) or a removable UICC commonly known as ‘SIM card.’ The memory QQ210 may allow the UE QQ200 to access instructions, application programs and the like, stored on transitory or non-transitory memory media, to off-load data, or to upload data. An article of manufacture, such as one utilizing a communication system may be tangibly embodied as or in the memory QQ210, which may be or comprise a device-readable storage medium.

[0382] The processing circuitry QQ202 may be configured to communicate with an access network or other network using the communication interface QQ212. The communication interface QQ212 may comprise one or more communication subsystems and may include or be communicatively coupled to an antenna QQ222. The communication interface QQ212 may include one or more transceivers used to communicate, such as by communicating with one or more remote transceivers of another device capable of wireless communication (e.g., another UE or a network node in an access network). Each transceiver may include a transmitter QQ218 and / or a receiver QQ220 appropriate to provide network communications (e.g., optical, electrical, frequency allocations, and so forth). Moreover, the transmitter QQ218 and receiver QQ220 may be coupled to one or more antennas (e.g., antenna QQ222) and may share circuit components, software or firmware, or alternatively be implemented separately.

[0383] In the illustrated embodiment, communication functions of the communication interface QQ212 may include cellular communication, Wi-Fi communication, LPWAN communication, data communication, voice communication, multimedia communication, short-range communications such as Bluetooth, near-field communication, location-based communication such as the use of the global positioning system (GPS) to determine a location, another like communication function, or any combination thereof. Communications may be implemented in according to one or more communication protocols and / or standards, such as IEEE 802.11, Code Division Multiplexing Access (CDMA), Wideband Code Division Multiple Access (WCDMA), GSM, LTE, New Radio (NR), UMTS, WiMax, Ethernet, transmission control protocol / internet protocol (TCP / IP), synchronous optical networking (SONET), Asynchronous Transfer Mode (ATM), QUIC, Hypertext Transfer Protocol (HTTP), and so forth.

[0384] Regardless of the type of sensor, a UE may provide an output of data captured by its sensors, through its communication interface QQ212, via a wireless connection to a network node. Data captured by sensors of a UE can be communicated through a wireless connection to a network node via another UE. The output may be periodic (e.g., once every 15 minutes if it reports the sensed temperature), random (e.g., to even out the load from reporting from several sensors), in response to a triggering event (e.g., when moisture is detected an alert is sent), in response to a request (e.g., a user initiated request), or a continuous stream (e.g., a live video feed of a patient).

[0385] As another example, a UE comprises an actuator, a motor, or a switch, related to a communication interface configured to receive wireless input from a network node via a wireless connection. In response to the received wireless input the states of the actuator, the motor, or the switch may change. For example, the UE may comprise a motor that adjusts the control surfaces or rotors of a drone in flight according to the received input or to a robotic arm performing a medical procedure according to the received input.

[0386] A UE, when in the form of an Internet of Things (loT) device, may be a device for use in one or more application domains, these domains comprising, but not limited to, city wearable technology, extended industrial application and healthcare. Non-limiting examples of such an loT device are a device which is or which is embedded in: a connected refrigerator or freezer, a TV, a connected lighting device, an electricity meter, a robot vacuum cleaner, a voice controlled smart speaker, a home security camera, a motion detector, a thermostat, a smoke detector, a door / window sensor, a flood / moisture sensor, an electrical door lock, a connected doorbell, an air conditioning system like a heat pump, an autonomous vehicle, a surveillance system, a weather monitoring device, a vehicle parking monitoring device, an electric vehicle charging station, a smart watch, a fitness tracker, a head-mounted display for Augmented Reality (AR) or Virtual Reality (VR), a wearable for tactile augmentation or sensory enhancement, a water sprinkler, an animal- or item-tracking device, a sensor for monitoring a plant or animal, an industrial robot, an Unmanned Aerial Vehicle (UAV), and any kind of medical device, like a heart rate monitor or a remote controlled surgical robot. A UE in the form of an loT device comprises circuitry and / or software in dependence of the intended application of the loT device in addition to other components as described in relation to the UE QQ200 shown in Figure QQ2.

[0387] As yet another specific example, in an loT scenario, a UE may represent a machine or other device that performs monitoring and / or measurements, and transmits the results of such monitoring and / or measurements to another UE and / or a network node. The UE may in this case be an M2M device, which may in a 3GPP context be referred to as an MTC device. As one particular example, the UE may implement the 3GPP NB-loT standard. In other scenarios, a UE may represent a vehicle, such as a car, a bus, a truck, a ship and an airplane, or other equipment that is capable of monitoring and / or reporting on its operational status or other functions associated with its operation.

[0388] In practice, any number of UEs may be used together with respect to a single use case. For example, a first UE might be or be integrated in a drone and provide the drone’s speed information (obtained through a speed sensor) to a second UE that is a remote controller operating the drone. When the user makes changes from the remote controller, the first UE may adjust the throttle on the drone (e.g. by controlling an actuator) to increase or decrease the drone’s speed. The first and / or the second UE can also include more than one of the functionalities described above. For example, a UE might comprise the sensor and the actuator, and handle communication of data for both the speed sensor and the actuators.

[0389] Figure 17 shows a network node QQ300 in accordance with some embodiments. As used herein, network node refers to equipment capable, configured, arranged and / or operable to communicate directly or indirectly with a UE and / or with other network nodes or equipment, in a telecommunication network. Examples of network nodes include, but are not limited to, access points (APs) (e.g., radio access points), base stations (BSs) (e.g., radio base stations, Node Bs, evolved Node Bs (eNBs) and NR NodeBs (gNBs)), O- RAN nodes or components of an O-RAN node (e.g., 0-Rll, 0-Dll, O-CU).

[0390] Base stations may be categorized based on the amount of coverage they provide (or, stated differently, their transmit power level) and so, depending on the provided amount of coverage, may be referred to as femto base stations, pico base stations, micro base stations, or macro base stations. A base station may be a relay node or a relay donor node controlling a relay. A network node may also include one or more (or all) parts of a distributed radio base station such as centralized digital units, distributed units (e.g., in an O-RAN access node) and / or remote radio units (RRUs), sometimes referred to as Remote Radio Heads (RRHs). Such remote radio units may or may not be integrated with an antenna as an antenna integrated radio. Parts of a distributed radio base station may also be referred to as nodes in a distributed antenna system (DAS).

[0391] Other examples of network nodes include multiple transmission point (multi-TRP) 5G access nodes, multi-standard radio (MSR) equipment such as MSR BSs, network controllers such as radio network controllers (RNCs) or base station controllers (BSCs), base transceiver stations (BTSs), transmission points, transmission nodes, multi- cel l / multicast coordination entities (MCEs), Operation and Maintenance (O&M) nodes, Operations Support System (OSS) nodes, Self-Organizing Network (SON) nodes, positioning nodes (e.g., Evolved Serving Mobile Location Centers (E-SMLCs)), and / or Minimization of Drive Tests (MDTs).

[0392] The network node QQ300 includes a processing circuitry QQ302, a memory QQ304, a communication interface QQ306, and a power source QQ308. The network node QQ300 may be composed of multiple physically separate components (e.g., a NodeB component and a RNC component, or a BTS component and a BSC component, etc.), which may each have their own respective components. In certain scenarios in which the network node QQ300 comprises multiple separate components (e.g., BTS and BSC components), one or more of the separate components may be shared among several network nodes. For example, a single RNC may control multiple NodeBs. In such a scenario, each unique NodeB and RNC pair, may in some instances be considered a single separate network node. In some embodiments, the network node QQ300 may be configured to support multiple radio access technologies (RATs). In such embodiments, some components may be duplicated (e.g., separate memory QQ304 for different RATs) and some components may be reused (e.g., a same antenna QQ310 may be shared by different RATs). The network node QQ300 may also include multiple sets of the various illustrated components for different wireless technologies integrated into network node QQ300, for example GSM, WCDMA, LTE, NR, WiFi, Zigbee, Z-wave, LoRaWAN, Radio Frequency Identification (RFID) or Bluetooth wireless technologies. These wireless technologies may be integrated into the same or different chip or set of chips and other components within network node QQ300.

[0393] The processing circuitry QQ302 may comprise a combination of one or more of a microprocessor, controller, microcontroller, central processing unit, digital signal processor, application-specific integrated circuit, field programmable gate array, or any other suitable computing device, resource, or combination of hardware, software and / or encoded logic operable to provide, either alone or in conjunction with other network node QQ300 components, such as the memory QQ304, to provide network node QQ300 functionality.

[0394] In some embodiments, the processing circuitry QQ302 includes a system on a chip (SOC). In some embodiments, the processing circuitry QQ302 includes one or more of radio frequency (RF) transceiver circuitry QQ312 and baseband processing circuitry QQ314. In some embodiments, the radio frequency (RF) transceiver circuitry QQ312 and the baseband processing circuitry QQ314 may be on separate chips (or sets of chips), boards, or units, such as radio units and digital units. In alternative embodiments, part or all of RF transceiver circuitry QQ312 and baseband processing circuitry QQ314 may be on the same chip or set of chips, boards, or units.

[0395] The memory QQ304 may comprise any form of volatile or non-volatile computer- readable memory including, without limitation, persistent storage, solid-state memory, remotely mounted memory, magnetic media, optical media, random access memory (RAM), read-only memory (ROM), mass storage media (for example, a hard disk), removable storage media (for example, a flash drive, a Compact Disk (CD) or a Digital Video Disk (DVD)), and / or any other volatile or non-volatile, non-transitory device- readable and / or computer-executable memory devices that store information, data, and / or instructions that may be used by the processing circuitry QQ302. The memory QQ304 may store any suitable instructions, data, or information, including a computer program, software, an application including one or more of logic, rules, code, tables, and / or other instructions capable of being executed by the processing circuitry QQ302 and utilized by the network node QQ300. The memory QQ304 may be used to store any calculations made by the processing circuitry QQ302 and / or any data received via the communication interface QQ306. In some embodiments, the processing circuitry QQ302 and memory QQ304 is integrated. The communication interface QQ306 is used in wired or wireless communication of signaling and / or data between a network node, access network, and / or UE. As illustrated, the communication interface QQ306 comprises port(s) / terminal(s) QQ316 to send and receive data, for example to and from a network over a wired connection. The communication interface QQ306 also includes radio front-end circuitry QQ318 that may be coupled to, or in certain embodiments a part of, the antenna QQ310. Radio front-end circuitry QQ318 comprises filters QQ320 and amplifiers QQ322. The radio front-end circuitry QQ318 may be connected to an antenna QQ310 and processing circuitry QQ302. The radio front-end circuitry may be configured to condition signals communicated between antenna QQ310 and processing circuitry QQ302. The radio front-end circuitry QQ318 may receive digital data that is to be sent out to other network nodes or UEs via a wireless connection. The radio front-end circuitry QQ318 may convert the digital data into a radio signal having the appropriate channel and bandwidth parameters using a combination of filters QQ320 and / or amplifiers QQ322. The radio signal may then be transmitted via the antenna QQ310. Similarly, when receiving data, the antenna QQ310 may collect radio signals which are then converted into digital data by the radio front-end circuitry QQ318. The digital data may be passed to the processing circuitry QQ302. In other embodiments, the communication interface may comprise different components and / or different combinations of components.

[0396] In certain alternative embodiments, the network node QQ300 does not include separate radio front-end circuitry QQ318, instead, the processing circuitry QQ302 includes radio front-end circuitry and is connected to the antenna QQ310. Similarly, in some embodiments, all or some of the RF transceiver circuitry QQ312 is part of the communication interface QQ306. In still other embodiments, the communication interface QQ306 includes one or more ports or terminals QQ316, the radio front-end circuitry QQ318, and the RF transceiver circuitry QQ312, as part of a radio unit (not shown), and the communication interface QQ306 communicates with the baseband processing circuitry QQ314, which is part of a digital unit (not shown).

[0397] The antenna QQ310 may include one or more antennas, or antenna arrays, configured to send and / or receive wireless signals. The antenna QQ310 may be coupled to the radio front-end circuitry QQ318 and may be any type of antenna capable of transmitting and receiving data and / or signals wirelessly. In certain embodiments, the antenna QQ310 is separate from the network node QQ300 and connectable to the network node QQ300 through an interface or port. The antenna QQ310, communication interface QQ306, and / or the processing circuitry QQ302 may be configured to perform any receiving operations and / or certain obtaining operations described herein as being performed by the network node. Any information, data and / or signals may be received from a UE, another network node and / or any other network equipment. Similarly, the antenna QQ310, the communication interface QQ306, and / or the processing circuitry QQ302 may be configured to perform any transmitting operations described herein as being performed by the network node. Any information, data and / or signals may be transmitted to a UE, another network node and / or any other network equipment.

[0398] The power source QQ308 provides power to the various components of network node QQ300 in a form suitable for the respective components (e.g., at a voltage and current level needed for each respective component). The power source QQ308 may further comprise, or be coupled to, power management circuitry to supply the components of the network node QQ300 with power for performing the functionality described herein. For example, the network node QQ300 may be connectable to an external power source (e.g., the power grid, an electricity outlet) via an input circuitry or interface such as an electrical cable, whereby the external power source supplies power to power circuitry of the power source QQ308. As a further example, the power source QQ308 may comprise a source of power in the form of a battery or battery pack which is connected to, or integrated in, power circuitry. The battery may provide backup power should the external power source fail.

[0399] Embodiments of the network node QQ300 may include additional components beyond those shown in Figure 17 for providing certain aspects of the network node’s functionality, including any of the functionality described herein and / or any functionality necessary to support the subject matter described herein. For example, the network node QQ300 may include user interface equipment to allow input of information into the network node QQ300 and to allow output of information from the network node QQ300. This may allow a user to perform diagnostic, maintenance, repair, and other administrative functions for the network node QQ300.

[0400] Figure 18 is a block diagram of a host QQ400, which may be an embodiment of the host QQ116 of Figure QQ1, in accordance with various aspects described herein. As used herein, the host QQ400 may be or comprise various combinations hardware and / or software, including a standalone server, a blade server, a cloud-implemented server, a distributed server, a virtual machine, container, or processing resources in a server farm. The host QQ400 may provide one or more services to one or more UEs.

[0401] The host QQ400 includes processing circuitry QQ402 that is operatively coupled via a bus QQ404 to an input / output interface QQ406, a network interface QQ408, a power source QQ410, and a memory QQ412. Other components may be included in other embodiments. Features of these components may be substantially similar to those described with respect to the devices of previous figures, such as Figures 14 and QQ3, such that the descriptions thereof are generally applicable to the corresponding components of host QQ400.

[0402] The memory QQ412 may include one or more computer programs including one or more host application programs QQ414 and data QQ416, which may include user data, e.g., data generated by a UE for the host QQ400 or data generated by the host QQ400 for a UE. Embodiments of the host QQ400 may utilize only a subset or all of the components shown. The host application programs QQ414 may be implemented in a container-based architecture and may provide support for video codecs (e.g., Versatile Video Coding (VVC), High Efficiency Video Coding (HEVC), Advanced Video Coding (AVC), MPEG, VP9) and audio codecs (e.g., FLAG, Advanced Audio Coding (AAC), MPEG, G.711), including transcoding for multiple different classes, types, or implementations of UEs (e.g., handsets, desktop computers, wearable display systems, heads-up display systems). The host application programs QQ414 may also provide for user authentication and licensing checks and may periodically report health, routes, and content availability to a central node, such as a device in or on the edge of a core network. Accordingly, the host QQ400 may select and / or indicate a different host for over-the-top services for a UE. The host application programs QQ414 may support various protocols, such as the HTTP Live Streaming (HLS) protocol, Real-Time Messaging Protocol (RTMP), Real-Time Streaming Protocol (RTSP), Dynamic Adaptive Streaming over HTTP (MPEG-DASH), etc.

[0403] Figure 19 is a block diagram illustrating a virtualization environment QQ500 in which functions implemented by some embodiments may be virtualized. In the present context, virtualizing means creating virtual versions of apparatuses or devices which may include virtualizing hardware platforms, storage devices and networking resources. As used herein, virtualization can be applied to any device described herein, or components thereof, and relates to an implementation in which at least a portion of the functionality is implemented as one or more virtual components. Some or all of the functions described herein may be implemented as virtual components executed by one or more virtual machines (VMs) implemented in one or more virtual environments QQ500 hosted by one or more of hardware nodes, such as a hardware computing device that operates as a network node, UE, core network node, or host. Further, in embodiments in which the virtual node does not require radio connectivity (e.g., a core network node or host), then the node may be entirely virtualized. In some embodiments, the virtualization environment QQ500 includes components defined by the O-RAN Alliance, such as an O- Cloud environment orchestrated by a Service Management and Orchestration Framework via an 0-2 interface.

[0404] Applications QQ502 (which may alternatively be called software instances, virtual appliances, network functions, virtual nodes, virtual network functions, etc.) are run in the virtualization environment Q400 to implement some of the features, functions, and / or benefits of some of the embodiments disclosed herein.

[0405] Hardware QQ504 includes processing circuitry, memory that stores software and / or instructions executable by hardware processing circuitry, and / or other hardware devices as described herein, such as a network interface, input / output interface, and so forth. Software may be executed by the processing circuitry to instantiate one or more virtualization layers QQ506 (also referred to as hypervisors or virtual machine monitors (VMMs)), provide VMs QQ508a and QQ508b (one or more of which may be generally referred to as VMs QQ508), and / or perform any of the functions, features and / or benefits described in relation with some embodiments described herein. The virtualization layer QQ506 may present a virtual operating platform that appears like networking hardware to the VMs QQ508.

[0406] The VMs QQ508 comprise virtual processing, virtual memory, virtual networking or interface and virtual storage, and may be run by a corresponding virtualization layer QQ506. Different embodiments of the instance of a virtual appliance QQ502 may be implemented on one or more of VMs QQ508, and the implementations may be made in different ways. Virtualization of the hardware is in some contexts referred to as network function virtualization (NFV). NFV may be used to consolidate many network equipment types onto industry standard high volume server hardware, physical switches, and physical storage, which can be located in data centers, and customer premise equipment.

[0407] In the context of NFV, a VM QQ508 may be a software implementation of a physical machine that runs programs as if they were executing on a physical, non-virtualized machine. Each of the VMs QQ508, and that part of hardware QQ504 that executes that VM, be it hardware dedicated to that VM and / or hardware shared by that VM with others of the VMs, forms separate virtual network elements. Still in the context of NFV, a virtual network function is responsible for handling specific network functions that run in one or more VMs QQ508 on top of the hardware QQ504 and corresponds to the application QQ502.

[0408] Hardware QQ504 may be implemented in a standalone network node with generic or specific components. Hardware QQ504 may implement some functions via virtualization. Alternatively, hardware QQ504 may be part of a larger cluster of hardware (e.g. such as in a data center or CPE) where many hardware nodes work together and are managed via management and orchestration QQ510, which, among others, oversees lifecycle management of applications QQ502. In some embodiments, hardware QQ504 is coupled to one or more radio units that each include one or more transmitters and one or more receivers that may be coupled to one or more antennas. Radio units may communicate directly with other hardware nodes via one or more appropriate network interfaces and may be used in combination with the virtual components to provide a virtual node with radio capabilities, such as a radio access node or a base station. In some embodiments, some signaling can be provided with the use of a control system QQ512 which may alternatively be used for communication between hardware nodes and radio units.

[0409] Figure 20 shows a communication diagram of a host QQ602 communicating via a network node QQ604 with a UE QQ606 over a partially wireless connection in accordance with some embodiments. Example implementations, in accordance with various embodiments, of the UE (such as a UE QQ112a of Figure 15 and / or UE QQ200 of Figure QQ2), network node (such as network node QQ110a of Figure 15 and / or network node QQ300 of Figure QQ3), and host (such as host QQ116 of Figure 15 and / or host QQ400 of Figure QQ4) discussed in the preceding paragraphs will now be described with reference to Figure QQ6.

[0410] Like host QQ400, embodiments of host QQ602 include hardware, such as a communication interface, processing circuitry, and memory. The host QQ602 also includes software, which is stored in or accessible by the host QQ602 and executable by the processing circuitry. The software includes a host application that may be operable to provide a service to a remote user, such as the UE QQ606 connecting via an over-the-top (OTT) connection QQ650 extending between the UE QQ606 and host QQ602. In providing the service to the remote user, a host application may provide user data which is transmitted using the OTT connection QQ650. The network node QQ604 includes hardware enabling it to communicate with the host QQ602 and UE QQ606. The connection QQ660 may be direct or pass through a core network (like core network QQ106 of Figure QQ1) and / or one or more other intermediate networks, such as one or more public, private, or hosted networks. For example, an intermediate network may be a backbone network or the Internet.

[0411] The UE QQ606 includes hardware and software, which is stored in or accessible by UE QQ606 and executable by the UE’s processing circuitry. The software includes a client application, such as a web browser or operator-specific “app” that may be operable to provide a service to a human or non-human user via UE QQ606 with the support of the host QQ602. In the host QQ602, an executing host application may communicate with the executing client application via the OTT connection QQ650 terminating at the UE QQ606 and host QQ602. In providing the service to the user, the UE's client application may receive request data from the host's host application and provide user data in response to the request data. The OTT connection QQ650 may transfer both the request data and the user data. The UE's client application may interact with the user to generate the user data that it provides to the host application through the OTT connection QQ650.

[0412] The OTT connection QQ650 may extend via a connection QQ660 between the host QQ602 and the network node QQ604 and via a wireless connection QQ670 between the network node QQ604 and the UE QQ606 to provide the connection between the host QQ602 and the UE QQ606. The connection QQ660 and wireless connection QQ670, over which the OTT connection QQ650 may be provided, have been drawn abstractly to illustrate the communication between the host QQ602 and the UE QQ606 via the network node QQ604, without explicit reference to any intermediary devices and the precise routing of messages via these devices.

[0413] As an example of transmitting data via the OTT connection QQ650, in step QQ608, the host QQ602 provides user data, which may be performed by executing a host application. In some embodiments, the user data is associated with a particular human user interacting with the UE QQ606. In other embodiments, the user data is associated with a UE QQ606 that shares data with the host QQ602 without explicit human interaction. In step QQ610, the host QQ602 initiates a transmission carrying the user data towards the UE QQ606. The host QQ602 may initiate the transmission responsive to a request transmitted by the UE QQ606. The request may be caused by human interaction with the UE QQ606 or by operation of the client application executing on the UE QQ606. The transmission may pass via the network node QQ604, in accordance with the teachings of the embodiments described throughout this disclosure. Accordingly, in step QQ612, the network node QQ604 transmits to the UE QQ606 the user data that was carried in the transmission that the host QQ602 initiated, in accordance with the teachings of the embodiments described throughout this disclosure. In step QQ614, the UE QQ606 receives the user data carried in the transmission, which may be performed by a client application executed on the UE QQ606 associated with the host application executed by the host QQ602.

[0414] In some examples, the UE QQ606 executes a client application which provides user data to the host QQ602. The user data may be provided in reaction or response to the data received from the host QQ602. Accordingly, in step QQ616, the UE QQ606 may provide user data, which may be performed by executing the client application. In providing the user data, the client application may further consider user input received from the user via an input / output interface of the UE QQ606. Regardless of the specific manner in which the user data was provided, the UE QQ606 initiates, in step QQ618, transmission of the user data towards the host QQ602 via the network node QQ604. In step QQ620, in accordance with the teachings of the embodiments described throughout this disclosure, the network node QQ604 receives user data from the UE QQ606 and initiates transmission of the received user data towards the host QQ602. In step QQ622, the host QQ602 receives the user data carried in the transmission initiated by the UE QQ606.

[0415] One or more of the various embodiments improve the performance of OTT services provided to the UE QQ606 using the OTT connection QQ650, in which the wireless connection QQ670 forms the last segment.

[0416] In an example scenario, factory status information may be collected and analyzed by the host QQ602. As another example, the host QQ602 may process audio and video data which may have been retrieved from a UE for use in creating maps. As another example, the host QQ602 may collect and analyze real-time data to assist in controlling vehicle congestion (e.g., controlling traffic lights). As another example, the host QQ602 may store surveillance video uploaded by a UE. As another example, the host QQ602 may store or control access to media content such as video, audio, VR or AR which it can broadcast, multicast or unicast to UEs. As other examples, the host QQ602 may be used for energy pricing, remote control of non-time critical electrical load to balance power generation needs, location services, presentation services (such as compiling diagrams etc. from data collected from remote devices), or any other function of collecting, retrieving, storing, analyzing and / or transmitting data. In some examples, a measurement procedure may be provided for the purpose of monitoring data rate, latency and other factors on which the one or more embodiments improve. There may further be an optional network functionality for reconfiguring the OTT connection QQ650 between the host QQ602 and UE QQ606, in response to variations in the measurement results. The measurement procedure and / or the network functionality for reconfiguring the OTT connection may be implemented in software and hardware of the host QQ602 and / or UE QQ606. In some embodiments, sensors (not shown) may be deployed in or in association with other devices through which the OTT connection QQ650 passes; the sensors may participate in the measurement procedure by supplying values of the monitored quantities exemplified above, or supplying values of other physical quantities from which software may compute or estimate the monitored quantities. The reconfiguring of the OTT connection QQ650 may include message format, retransmission settings, preferred routing etc.; the reconfiguring need not directly alter the operation of the network node QQ604. Such procedures and functionalities may be known and practiced in the art. In certain embodiments, measurements may involve proprietary UE signaling that facilitates measurements of throughput, propagation times, latency and the like, by the host QQ602. The measurements may be implemented in that software causes messages to be transmitted, in particular empty or ‘dummy’ messages, using the OTT connection QQ650 while monitoring propagation times, errors, etc.

[0417] Although the computing devices described herein (e.g., UEs, network nodes, hosts) may include the illustrated combination of hardware components, other embodiments may comprise computing devices with different combinations of components. It is to be understood that these computing devices may comprise any suitable combination of hardware and / or software needed to perform the tasks, features, functions and methods disclosed herein. Determining, calculating, obtaining or similar operations described herein may be performed by processing circuitry, which may process information by, for example, converting the obtained information into other information, comparing the obtained information or converted information to information stored in the network node, and / or performing one or more operations based on the obtained information or converted information, and as a result of said processing making a determination. Moreover, while components are depicted as single boxes located within a larger box, or nested within multiple boxes, in practice, computing devices may comprise multiple different physical components that make up a single illustrated component, and functionality may be partitioned between separate components. For example, a communication interface may be configured to include any of the components described herein, and / or the functionality of the components may be partitioned between the processing circuitry and the communication interface. In another example, non-computationally intensive functions of any of such components may be implemented in software or firmware and computationally intensive functions may be implemented in hardware.

[0418] In certain embodiments, some or all of the functionality described herein may be provided by processing circuitry executing instructions stored on in memory, which in certain embodiments may be a computer program product in the form of a non-transitory computer-readable storage medium. In alternative embodiments, some or all of the functionality may be provided by the processing circuitry without executing instructions stored on a separate or discrete device-readable storage medium, such as in a hard-wired manner. In any of those particular embodiments, whether executing instructions stored on a non-transitory computer-readable storage medium or not, the processing circuitry can be configured to perform the described functionality. The benefits provided by such functionality are not limited to the processing circuitry alone or to other components of the computing device, but are enjoyed by the computing device as a whole, and / or by end users and a wireless network generally.

[0419] When using the word "comprise" or “comprising” it shall be interpreted as nonlimiting, i.e. meaning "consist at least of".

[0420] It will be appreciated that the foregoing description and the accompanying drawings represent non-limiting examples of the methods and apparatus taught herein. As such, the apparatus and techniques taught herein are not limited by the foregoing description and accompanying drawings. Instead, the embodiments herein are limited only by the following claims and their legal equivalents.

[0421] Abbreviation Explanation

[0422] ADAEC Application Data Analytics Enabler Client

[0423] ADAES Application Data Analytics Enabler Server

[0424] Al Artificial Intelligence

[0425] AnLF Analytics logical function

[0426] DML Distributed Machine Learning

[0427] FL Federated Learning

[0428] ML Machine Learning

[0429] MTLF Model Training logical function MTME Machine Learning Model Training and Management Enablement

[0430] MTMEC MTME Client

[0431] MTMES MTME Server

[0432] NWDAF Network Data Analytics Function SEAL Service Enabler Architecture Layer for Verticals

[0433] 5GC 5G Core Network

Claims

CLAIMS1. A method performed by a managing node (110) for training and / or managing a machine learning, ML, model associated with a vertical application layer, VAL, service in a communications network (100), the method comprising:- managing (201) one or more operations in the communications network, wherein the one or more operations are related to the ML model; and- executing (202) at least one out of the managed one or more operations in the communications network.

2. The method according to claim 1 , wherein managing the one or more or operation comprises any one or more out of:- managing and / or coordinating the training of the ML model,- managing and / or coordinating the distribution of the ML model,- managing and / or coordinating the transfer of the ML model, training and / or inference,- managing and / or coordinating the interactions among one or more out of application servers, application clients, network functions, operation and maintenance nodes, and / or other network entities for the ML operations, and- managing and / or coordinating incoming requests for ML model training and / or ML management from a requesting node (130).

3. The method according to claim 2, wherein managing and / or coordinating the training of the ML model comprises any one or more out of:- discovering Distributed ML, DML, Federated Learning, FL, and / or Split Learning group members,- selecting DML, FL and / or Split Learning group members,- exchanging ML models, ML model parameters and / or training data among selected DML, FL and / or Split Learning group members, wherein the ML models ML model parameters and / or training data comprises any one or more out of initial, intermediate and final ML models, ML model parameters and / or training data,- monitoring the training process and model performance, and- maintaining the training process and DML, FL and / or Split Learning group members when dynamic changes of status among the DML, FL and / or Split Learning group members occurs.

4. The method according to any of claims 1-3, wherein executing the at least one managed operation comprises any one or more out of:- executing the training of ML model relevant operations,- executing the distribution of ML model, and- executing the transfer of ML model, training and / or inference.

5. The method according to any of claims 1-4, wherein the managing node (110) is as any one out of:- a standalone Service Enabler Analytics Layer for Verticals, SEAL, function.- collocated with an existing SEAL function, or- a new enabler layer.

6. A computer program (1230) comprising instructions, which when executed by a processor (1210), causes the processor (1210) to perform actions according to any of claims 1-5.

7. A carrier (1240) comprising the computer program (1230) of claim 6, wherein the carrier (1240) is one of an electronic signal, an optical signal, an electromagnetic signal, a magnetic signal, an electric signal, a radio signal, a microwave signal, or a computer- readable storage medium.

8. A method performed by an executing node (121) for training a machine learning, ML, model associated with a vertical application layer, VAL, service in a communications network (100), the method comprising:- executing (301) one or more operations in the communications network, wherein the one or more operations are related to training the ML model.

9. The method according to claim 8, wherein executing the at least one managed operation comprises any one or more out of:- executing the training of ML model relevant operations,- executing the distribution of ML model, and- executing the transfer of ML model, training and / or inference.

10. The method according to any of claims 8-9, wherein the executing node (121) is any one out of:- a standalone Service Enabler Analytics Layer for Verticals, SEAL, function.- collocated with an existing SEAL function, or- a new enabler layer.

11. A computer program (1330) comprising instructions, which when executed by a processor (1310), causes the processor (1310) to perform actions according to any of claims 8-10.

12. A carrier (1340) comprising the computer program (1330) of claim 11 , wherein the carrier (1340) is one of an electronic signal, an optical signal, an electromagnetic signal, a magnetic signal, an electric signal, a radio signal, a microwave signal, or a computer- readable storage medium.

13. A method performed by a requesting node (130) for training, managing and / or providing a machine learning, ML, model associated with a vertical application layer, VAL, service in a communications network (100), the method comprising:- requesting (401) a managing node (110) and / or an executing node (121) to perform one or more operations in the communications network, wherein the one or more operations are related to managing, training and / or providing the ML model.

14. The method according to claim 13, wherein the one or more operation comprises any one or more out of:- training and / or providing the ML model,- managing distribution of the ML model, and- managing transfer for the ML model.

15. The method according to any of clams 13-14 wherein the method further comprises:- receiving (402) a notification from the managing node (110) and / or the executing node (121), the notification notifying the requesting node (130) that the training is completed, and / or comprising the trained ML model.

16. The method according to any of claims 13-15, wherein the requesting (401) further comprises receiving a response from the managing node (110) and / or executing node (121), the response acknowledging the request.

17. A computer program (1430) comprising instructions, which when executed by a processor (1410), causes the processor (1410) to perform actions according to any of claims 13-16.

18. A carrier (1440) comprising the computer program (1430) of claim 17, wherein the carrier (1440) is one of an electronic signal, an optical signal, an electromagnetic signal, a magnetic signal, an electric signal, a radio signal, a microwave signal, or a computer- readable storage medium.

19. A managing node (110) configured to train and / or manage a machine learning, ML, model associated with a vertical application layer, VAL, service in a communications network (100), the managing node (121) further being configured to:- manage one or more operations in the communications network, wherein the one or more operations are adapted to be related to the ML model; and- execute at least one out of the managed one or more operations in the communications network.

20. The managing node (110) according to claim 19, wherein the managing node (110) is configured to manage the one or more or operation by further being figured to any one or more out of:- manage and / or coordinate the training of the ML model,- manage and / or coordinate the distribution of the ML model,- manage and / or coordinate the transfer of the ML model, training and / or inference,- manage and / or coordinate the interactions among one or more out of application servers, application clients, network functions, operation and maintenance nodes, and / or other network entities for the ML operations, and- manage and / or coordinate incoming requests for ML model training and / or ML management from a requesting node (130).

21. The managing node (110) according to claim 20, wherein manage and / or coordinate the training of the ML model is adapted to comprise any one or more out of:- discover Distributed ML, DML, Federated Learning, FL, and / or Split Learning group members,- select DML, FL and / or Split Learning group members,- exchange ML models, ML model parameters and / or training data among selected DML, FL and / or Split Learning group members, wherein the ML models ML model parameters and / or training data is adapted to comprise any one or more out of initial, intermediate and final ML models, ML model parameters and / or training data,- monitor the training process and model performance, and- maintain the training process and DML, FL and / or Split Learning group members when dynamic changes of status among the DML, FL and / or Split Learning group members occurs.

22. The managing node (110) according to any of claims 19-21 , wherein the managing node (110) is configured to execute the at least one managed operation by further being configured to any one or more out of:- execute the training of ML model relevant operations,- execute the distribution of ML model, and- execute the transfer of ML model, training and / or inference.

23. The managing node (110) according to any of claims 19-22, wherein the managing node (110) adapted to be any one out of:- a standalone Service Enabler Analytics Layer for Verticals, SEAL, function.- collocated with an existing SEAL function, or- a new enabler layer.

24. An executing node (121) configured to train a machine learning, ML, model associated with a vertical application layer, VAL, service in a communications network (100), the executing node (121) further being configured to:- executing one or more operations in the communications network, wherein the one or more operations are adapted to be related to training the ML model.

25. The executing node (121) according to claim 24, wherein the executing node (121) is configured to execute the at least one managed operation by further being configured to any one or more out of:- execute the training of ML model relevant operations,- execute the distribution of ML model, and- execute the transfer of ML model, training and / or inference.

26. The executing node (121) according to any of claims 24-25, wherein the executing node (121) is adapted to be any one out of:- a standalone Service Enabler Analytics Layer for Verticals, SEAL, function.- collocated with an existing SEAL function, or- a new enabler layer.

27. A requesting node (130) configured to train, manage and / or provide a machine learning, ML, model associated with a vertical application layer, VAL, service in a communications network (100), the requesting node (130) further being configured to:- request a managing node (110) and / or an executing node (121) to perform one or more operations in the communications network, wherein the one or more operations are adapted to be related to managing, training and / or providing the ML model.

28. The requesting node (130) according to claim 27, wherein the one or more operation is adapted to comprise any one or more out of:- train and / or providing the ML model,- manage distribution of the ML model, and- manage transfer for the ML model.

29. The requesting node (130) according to any of clams 27-28 wherein the requesting node (130) is further configured to:- receive a notification from the managing node (110) and / or the executing node (121), the notification is adapted to notify the requesting node (130) that the training is completed, and / or comprising the trained ML model.

30. The requesting node (130) according to any of claims 27-29, wherein the requesting node (130) is configured to request the managing node (110) and / or the executing node (121) to perform one or more operations by further being configured to receive a response from the managing node (110) and / or executing node (121), the response acknowledging the request.

Images