Machine learning model management orchestration
By classifying queries and routing them to appropriate LLMs in a domain-specific manner, the method addresses latency and privacy issues in mobile networks, enhancing the performance and efficiency of ML model deployment.
Patent Information
- Application Number
- PCT/EP2025/074648
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2025-05-16
- Filing Date
- 2025-08-29
- Publication Date
- 2026-11-19
AI Technical Summary
Deploying large language models (LLMs) in mobile networks faces challenges due to latency and privacy concerns, especially when centralized models are cost-prohibitive, and fine-tuning smaller LLMs for domain-specific tasks is hindered by limited data availability.
A computer-implemented method for classifying user queries into domains and routing them to appropriate LLMs, allowing for fine-tuning and orchestration of ML models closer to the network edge, ensuring lower response times and data privacy.
This approach reduces latency and ensures data privacy by hosting ML models closer to the network edge, providing efficient and cost-effective domain-specific performance.
Smart Images

Figure EP2025074648_19112026_PF_FP_ABST
Abstract
Description
MACHINE LEARNING MODEL MANAGEMENT ORCHESTRATIONTECHNICAL FIELD
[0001] The present disclosure is related to computer-implemented methods and apparatus for machine learning (ML) model management orchestration in a telecommunication network, and related methods and apparatus.BACKGROUND
[0002] Some ML models including, for example, generative models such as large language models (LLMs) have exhibited remarkable capabilities in tasks such as question-answering (QA), solving mathematical, logical, or code-related problems, as well as creative types of problems. While size and computational demands may not be the only denominator of performance of an LLM, ML models with a greater number of parameters may tend to perform better than ML models with smaller number of parameters. For example, OpenAI's Research on scaling laws for neural language models describes that performance improves predictably with increases in model size, data, and compute. Jared Kaplan, et. al., Scaling Laws for Neural Language Models, arXiv:2001, 08361vl (23 Jan. 2020).
[0003] On the other hand, smaller ML models (e.g., such as those with fewer than 10 billion parameters), when fine-tuned on domain-specific datasets, may tend to outperform larger LLMs on highly specific questions in terms of accuracy, relevancy, and interpretability by a significant margin. See e.g., Pranjali Awasthi, Maximizing Use-Case Specificity through Precision Model Tuning, arXiv:2212.14206vl (submitted 29 Dec. 2022).SUMMARY
[0004] Challenges may occur when deploying a ML model, such as a LLM, for production in a mobile network. As the technology advances, the use cases for LLMs, for example, are anticipated to grow in the coming years, including applications in a radio access network (RAN), transport network, and / or core network. However, certain use cases may impose strict latency and privacy requirements for example, which can make it impractical to rely on LLMs with many parameters that are typically centralized due to cost constraints.
[0005] One approach may be to deploy smaller LLMs closer to the network edge that are finetuned for a specific set of queries and address the latency and privacy requirements in a cost-effective manner.
[0006] However, challenges may still exist regarding how to effectively fine-tune and align ML models, e.g., LLMs, to understand and address a specific task and requirements of a given domain, particularly in the presence of limited domain-specific data.
[0007] Certain aspects of the disclosure and their embodiments may provide solutions to these or other challenges. In some embodiments, a computer-implemented method is provided for classifying at least one user query to different domains and routing the at least one user query to a corresponding LLM. In some embodiments, in addition to performing routing, the computer-implemented method, can initiate performance of orchestration actions such as transferring a LLM(s) to other locations or orchestrating a new LLM(s). In addition to serving the needs of a mobile network operator, the computer-implemented method of some embodiments can be exposed and provided as-a-service for a third party, such as a third-party developer. The terms “query”, “user query”, or “prompt” are used herein in a non-limiting manner and may be interchangeable and replaced with the term “task”.
[0008] In some embodiments of the present disclosure, a computer implemented method for ML model management orchestration in a telecommunication network is provided. The computer-implemented method includes accessing at least one task directed towards at least one ML model in a plurality of ML models in the telecommunications network; and classifying the at least one task into one or more domains in a plurality of domains in the telecommunications network. The classifying is based on a classification of the at least one task into one or more domains to which the at least one task belongs. The computer-implemented method further includes routing the at least one task to a ML model in a domain based on the classification.
[0009] According to other embodiments, a node having a logical function is provided. The node includes processing circuitry, and memory coupled with the processing circuitry. The memory includes instructions that when executed by the processing circuitry causes the logical function to perform operations. The operations include to access at least one task directed towards at least one ML model in a plurality of ML models in the telecommunications network; and to classify the at least one task into one or more domains in a plurality of domains in the telecommunications network. The classifying is based on a classification of the at least one task into one or more domains to which the at least one task belongs. The operations further include to route the at least one task to a ML model in a domain based on the classification.
[0010] In other embodiments, a computer program, product including a non-transitory storage medium including program code to be executed by processing circuitry of a node is provided.Execution of the program code causes a logical function to perform operations. The operationsinclude to access at least one task directed towards at least one ML model in a plurality of ML models in the telecommunications network; and to classify the at least one task into one or more domains in a plurality of domains in the telecommunications network. The classifying is based on a classification of the at least one task into one or more domains to which the at least one task belongs. The operations further include to route the at least one task to a ML model in a domain based on the classification.
[0011] Certain embodiments may provide one or more of the following technical advantages. By classifying a task(s) into one or more domains to which the task(s) belongs and routing the task(s) to a ML model in the domain based on the classification, lower response times may be provided. In an example, the method results in hosting a ML model closer to the telecommunication network edge and, thus, response time is lower. Moreover, based on classifying a task(s) into one or more domains to which the task(s) belongs and routing the task(s) to a ML model in the domain based on the classification privacy of data may be ensured. Further, in certain embodiments the computer-implemented method is exposed to an external third party. As a consequence, distributed cloud providers or application functions, for example, may host the ML model(s).BRIEF DESCRIPTION OF THE DRAWINGS
[0012] The accompanying drawings, which are included to provide a further understanding of the disclosure and are incorporated in and constitute a part of this application, illustrate certain nonlimiting embodiments of inventive concepts. In the drawings:
[0013] Figures 1 A-1C are block diagrams illustrating an example of components and operations according to some embodiments;
[0014] Figure 2 is a sequence diagram illustrating operations in accordance with some embodiments;
[0015] Figure 3 is schematic drawing illustrating example domains for clustering in accordance with some embodiments;
[0016] Figures 4A and 4B are flow charts of operations of a node in accordance with some embodiments;
[0017] Figure 5 is a block diagram of a communication system in accordance with some embodiments;
[0018] Figures 6A and 6B are block diagrams illustrating an example of a node and logical function(s) and operations in accordance with some embodiments;
[0019] Figure 7 is a block diagram of a node in accordance with some embodiments; and
[0020] Figure 8 is a block diagram of a virtualization environment in accordance with some embodiments.DETAILED DESCRIPTION
[0021] 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, in which examples of embodiments of the present disclosure are shown. The present disclosure may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of present inventive concepts to those skilled in the art. It should also be noted that these embodiments are not mutually exclusive. Components from one embodiment may be tacitly assumed to be present / used in another embodiment.
[0022] While embodiments discussed herein are explained in the non-limiting context of a LLM(s), the disclosure is not so limited. Instead, the method can be used in other cases, including without limitation, other types of a ML model(s) whose size and computational demands affect performance of the ML model(s), and / or that have a large number of parameters.
[0023] The computer-implemented method of some embodiments can classify queries into specific domains and route them to a corresponding ML model(s). Additionally, the computer-implemented method can initiate orchestration of actions such as deploying a new ML model(s) or relocating an existing ML model(s) to optimize performance. Moreover, in some embodiments, these capabilities can be exposed to a third party, such as a third party developer.
[0024] In contrast, other approaches, such as NetOrchLLM for example, may integrate a LLM to enhance wireless network management without domain-based query routing and / or without exposing the approach to a third party. See e.g., Asmaa Abdallah, et. al., NetOrchLLM: Mastering Wireless Network Orchestration with Large Language Models, arXiv:2412.10107vl (submitted 13 Dec. 2024).
[0025] While another approach may include a LLM serving as a sixth generation (6G) orchestrator to automate physical-layer processes, such an approach does not include at least domain-based query routing or exposing the approach to a third party. See e.g., Zhuoran Xiao, et.al., “LLM Agents as 6G Orchestrator: A Paradigm for Task-Oriented Physical-Layer Automation, arXiv:2410.03688vl (submitted 21 Sep. 2024).
[0026] Caching approaches may include analysis of LLM tasks, such as semantic analysis, but do not address at least orchestration and / or optimization. See e.g., Jiaxing Li, et. al., “SCALM:Towards Semantic Caching for Automated Chat Services with Large Language Models, arXiv:2406.00025 (submitted 24 May 2024).
[0027] Some embodiments or the present disclosure include clustering of a task(s) per domain and area of origin, orchestration of a ML model(s) based on task clusters and available resources, and routing of a task(s) to an appropriate ML model(s).
[0028] Figures 1 A-1C are block diagrams illustrating an example of components and operations according to some embodiments. Figure 1 A illustrates an example of components and operations for creating a node profile. Figure IB illustrates an example of components and operations for clustering tasks. Figure 1C illustrates an example of components and operations for ML model orchestration.
[0029] As used herein, one or more “nodes” refers to entities that can submit tasks and / or host a ML model(s), such as a LLM(s), that can solve the tasks. A node can, for example, be located at a network edge (e.g., as part of, or adjacent to radio base stations such as fifth generation (5G) g node-Bs (gNBs); can be a more centralized function(s); and / or can be external to an operator’s network.
[0030] Figures 1 A- 1C illustrate a node that includes a model management function (MMF) 110. MMF 110 in this example include a node database 112 (node DB 112), task cluster(s) storage 114, and node topology storage 116. MMF 110 can route task descriptions to one or more of node 1 . . . node X hosting ML models.
[0031] MMF 110 also can operate in the background, gathering information from one or more of node 1 . . . node X about their availability, capabilities, and the tasks they transmit. This information can be stored in node DB 112. As shown in the example in Figure 1 A, using this data, MMF 110 creates a node profile(s) and a task description(s).
[0032] Referring to Figure IB, MMF 100 can then use the node profile(s) and task description(s) to cluster the task(s) by domain (e.g., RAN, transport network, core network, etc.). The task cluster(s) can be stored in task cluster storage 114.
[0033] Referring to Figure 1C, the task cluster(s), along with the node profile(s), are provided as input to a orchestration action creator 118 internal to MMF 110. In this example, the orchestration action creator 118 determines which ML model to deploy on which node X, based on a node topology graph from node topology storage 116.
[0034] Figure 2 is a sequence diagram illustrating operations of an example embodiment. A first background process includes operations 1-6, in which MMF 110 retrieves information from node X in operation 1 and performs operations 2-6. A second background process includes operations 7-11, in which MMF 110 retrieves information from node X in operation 7 and performsoperations 8-11. The retrieval in operations 1, 7 can occur periodically or asynchronously, such as when node X notifies MMF 110 of updates. The operations may be initiated by a MMF 110 request or follow a publish / subscribe model, where node X transmits information to MMF 110 without explicit requests.
[0035] In the first background process, a node profile(s) and a node relationship, e.g. a node topology in this non-limiting example, are created based on information provided from at least node X in operation 1. The information, which is used to create the node profile, is identified by a unique identifier (“node id”). The information can include:- Capacity, which refers to the maximum amount of work, data, or resources a node can handle or process. Capacity represents the upper limit of the node's performance capabilities. A node's capacity may be defined in terms of processing power (e.g., CPU cycles), memory, bandwidth, and / or the number of simultaneous connections it can support.- Load, which refers to the current amount of work, data, or resource utilization on the node.Load reflects the real-time usage relative to the node's capacity. For instance, load can represent the current processing demand, current energy consumption, and / or active user equipment (UE) connections, and / or data traffic being handled by the node- Location and mobility, which indicate the location of the node and the node’s movement in space.
[0036] The information can further include information about neighboring nodes (e.g., in terms of node id), and location information (e.g., in terms of physical location). In the example in Figure 2, the information is an information update that includes an update to one or more of the information.
[0037] In operation 2, an endpoint 200 of MMF 110 that received the information update, forwards the information update to node DB 112. Node DB 112, in operation 3, updates a node profile using the information update.
[0038] In operation 4, endpoint 200 forwards the information update to node topology 116. Node topology 116, in operation 5, uses the information update for the node topology. In the node topology, a respective node is represented by a unique node id and its physical location (e.g., latitude and longitude or Cartesian coordinates). The neighboring relationships between nodes are defined by the list of neighboring node id for each node. These details of the node topology can be stored in a data structure, like a dictionary, where each node’s information is mapped to its respective node id. Using this information, a network topology can be constructed, e.g. as a graph where nodes are vertices, and the edges represent the connections (or neighbors) between them. Thenode profile and node topology information, in this example, are timestamped at the time of extraction.
[0039] In operation 6, node DB 112 sends an acknowledgement (ACK) of the information update to endpoint 200.
[0040] In the second background process, in this example, a task(s) submitted by a node(s) at MMF 110 are clustered. The clustering is based on the domain the task belongs to. The domains can have a hierarchical split as shown in the example in Figure 3.
[0041] As shown in the example in Figure 3, example domains for task clusters for a mobile network 300 include infrastructure 310, management 350, and services 360.
[0042] Infrastructure 310 in this example includes RAN / base station 320, core network 330, and transport network 340. The RAN / base station 320, in turn, includes baseband board 320a, radio unit 320b, antenna 320c, power subsystem 320d, and cooling subsystem 320e. Core network 330, in turn, includes user plane functions 332 that include serving gateway (SGW) 332a and user plane function (UPF) 332b; and control plane functions 338 that include access and mobility function (AMF) 338a, unified data management (UDM) 338b, authentication server function (AUSF) 338, policy control function (PCF) 338d, and network exposure function (NEF) 338e. Transport network 340, in turn, includes microwave 340a, fiber 340b, and ethernet 340c.
[0043] Management 350 in this example includes operations support system (OSS) / business support system (BSS) 350a, open network automation platform (ONAP) 350b, and European Telecommunications Standards Institute management and orchestration (ETSI-MANO) 350c.
[0044] Services 360 in this example includes network slicing 360a, voice over long term evolution (VoLTE) / voice over new radio (VoNR) 360b, and content delivery network (CDN) 360c.
[0045] Tasks can belong to more than one cluster. Therefore, clustering incoming ML model task descriptions into predefined domains can be framed as a multi-label clustering task, where each task description can belong to one or multiple clusters (domains) based on its content.
[0046] Referring again to Figure 2, in the second background process, in operation 7, node X makes a task submission (e.g., a task description) to endpoint 200. Endpoint 200, in operation 8, forwards the task description to task clusters 114 for storage. Operations 9 and 10 are optionally performed if a certain number of tasks are submitted or a certain amount of time elapsed, or based on another condition such as a major outlier. In optional operation 9, task clusters 114 updates the node profile; and in operation 10, task clusters 114 sends an acknowledgement (ACK) of the update to endpoint 200. In operation 11, endpoint 200 forwards the task to node Y for processing (e.g., forwards node X address and task description).
[0047] To cluster the incoming task(s), in this example, the task description(s) is converted into a vector representation using techniques such as term frequency -inverse document frequency (TF-IDF), word embeddings (e.g., Word2Vec, GloVe), or transformer-based sentence embeddings (e.g., BERT). Then, a clustering algorithm such as K-Means, Density-Based Spatial Clustering (DBSCAN), or hierarchical clustering is used, where a task description can be assigned to more than one cluster, depending on its similarity to multiple centroids of clusters or clusters.
[0048] As shown in optional operations 9 and 10, MMR 110 can keep a temporary record of all incoming task descriptions and may decide to repeat clustering after a number of new incoming task descriptions after the last clustering or after an amount of time has elapsed or a combination thereof.
[0049] Still referring to Figure 2, an orchestration process includes operations 12-17. The orchestration process can be triggered by a variety of factors, for example after a given amount of incoming tasks have been received, after a given amount of time, or another event, for example after the mobile network observes degradation in the performance of the ML models hosted in one or more nodes. In the example in Figure 2, the orchestration process is handled by the orchestration action creator 118 component of MMF 110.
[0050] In operations 12-14, orchestration action creator 118 retrieves the task cluster(s) from task clusters 114, node profiles from node DB 112, and node relationships from node topology 116.
[0051] In operation 15, the orchestration action creator 118 uses the input received in operations 12-14 to propose one or more actions that are sent to the management and orchestration (MANO) 202 function of an underlying cloud infrastructure. These actions may involve at least one of the following: Create a new ML model instance of a given model ID at a specific node; transfer a ML model instance from one node to another node; distill a model instance to a simpler ML model using a model optimization technique, such as quantization or generative adversarial learning; and / or remove a ML model instance of a given model ID at a specific node.
[0052] In operations 16 and 17, orchestration action creator 118 forwards the chosen action or actions to MANO 202 via endpoint 200 of MMF 110. Endpoint 200 is a demarcation point between MMF 110 and a MANO 202 system that performs the orchestration action on the cloud infrastructure.
[0053] Figures 4A and 4B are flowcharts illustrating operations of a node having a logical function according to some embodiments.
[0054] Referring to Figure 4B, some embodiments are directed to a computer-implemented method for ML model management orchestration in a telecommunication network. The computer-implemented method includes accessing 408 at least one task directed towards at least one MLmodel in a plurality of ML models in the telecommunications network. The method further includes classifying 410 the at least one task into one or more domains in a plurality of domains in the telecommunications network. The classifying is based on a classification of the at least one task into one or more domains to which the at least one task belongs. The method further includes routing 412 the at least one task to a ML model in a domain based on the classification.
[0055] Referring to Figure 4A, in some embodiments, the computer-implemented method further includes accessing 400 node information from a plurality of nodes associated with the telecommunications network. The node information includes respective availability of respective nodes in the plurality of nodes, respective capabilities of the respective nodes, and at least one respective task directed by the respective nodes towards the telecommunication network. The method can further includes creating 402 a respective profile of a respective node and a respective task description of the at least one respective task. A respective profile can include at least an identifier of a respective node, a capacity of the respective node, and a load of the respective node.
[0056] In other embodiments, the computer-implemented method further includes creating 404 a description among the plurality of nodes. The description of the relationship among the plurality of nodes can include, per node, an identifier of a respective node, a location of the respective node, and neighboring relationships between nodes. In still other embodiments, the computer-implemented method can further include constructing 406 a node topology based on the description of the relationship among the plurality of nodes.
[0057] Classifying 410 can include clustering of tasks per domain, and the classification of the at least task belongs to one cluster comprising one domain. In other embodiments, classifying 410 includes clustering of tasks per domain. The classification of the at least one task can belong to a plurality of clusters, and respective clusters in the plurality of clusters respectively can comprise a different domain.
[0058] In still other embodiments, classifying 410 at least one respective task into one or more domains is based on using the node information to cluster the at least one respective task by domain. In some embodiments, classifying 410 at least one respective tasks includes clustering at least one respective task into one or more clusters based on the at least one respective task description of the at least one respective task.
[0059] In yet other embodiments, classifying 410 further includes clustering at least one respective task by area of origin in the telecommunications network of the at least respective task.
[0060] Clustering at least one respective task can include converting the at least one respective task description of the at least one respective task into a vector representation, and assigning the atleast one respective task to one or more clusters based on a similarity of the vector representation to a cluster or a centroid of a cluster.
[0061] In other embodiments, the computer-implemented method further includes repeating 414 classifying of at least one task into one or more domains based on at least one of (i) accessing a new task description, after a last classifying, that exceed a threshold value, and (ii) an elapsed time period.
[0062] Referring to Figure 4B, in still other embodiments, the computer-implemented method further includes initiating 416 at least one orchestration action for at least one ML model based on the classification of the at least one task into one or more domains. The initiating 416 can be further based on available resources of the one or more domains for the classified at least one task. In some embodiments, initiating 416 at least one orchestration action includes accessing information including the classification, information about a node from which the at least one task originated, and information about a relationship among a plurality of nodes associated with the telecommunications network; and determining one or more orchestration actions based on the accessed information.
[0063] In some embodiments, the at least one orchestration action includes at least one of: creating a new ML model instance of a given ML model identification at an identified node; transferring a ML model instance from a first node to a second node; distilling a ML model instance to a simpler ML model instance based on use of a ML model optimization; and removing a ML model instance of a given ML model identification at an identified node.
[0064] Initiating 416 at least one orchestration action can be triggered by at least one of an elapsed time period after accessing the at least one task, expiration of a time period, and an event in the telecommunication network.
[0065] A domain can include at least one of a RAN, a transport network, a core network, and / or at least one function of the RAN, the transport network, and the core network.
[0066] In some embodiments, the computer-implemented method further includes exposing 418 the computer-implemented method to an application function (AF).
[0067] In some embodiments, the ML model comprises a generative model.
[0068] Operations 400-406 of Figure 4 A and operations 414-418 of Figure 4B may be optional with respect to some embodiments of nodes and related methods.
[0069] Operations discussed with reference to the flow charts of Figures 4A and 4B according to some embodiments of the present disclosure may be performed by a node having a logical function (implemented using the structure of Figure 7 and / or Figure 8). For example, modules maybe stored in memory 704, logical functions 112, 114, 116, and / or 118 of Figure 7, and / or management and orchestration (MANO) component 810 of Figure 8, and these modules may provide instructions so that when the instructions of a module are executed by respective node processing circuitry 702, node 700, 800 performs respective operations of the flow charts.
[0070] Figure 5 shows an example of a communication system 500 in accordance with some embodiments.
[0071] In the example, the communication system 500 includes a telecommunications network 502 that includes an access network 504, such as RAN, and a core network 506, which includes one or more core network nodes 508. The access network 504 includes one or more access network nodes or base stations of various types, access network nodes 510A and 510B are depicted (which may be collectively referred to as network nodes 510), or any other similar 3rd Generation Partnership Project (3 GPP) access nodes or non-3GPP access points (APs). Some embodiments of the access network 504 may include more than one access network technology. The network nodes 510 of access network 504 facilitate direct or indirect connection of wireless devices, also referred to as user equipments (UEs), such as by connecting UEs 512A, 512B, 512C, and 512D (one or more of which may be generally referred to as UEs 512) to the core network 506 over one or more wireless connections.
[0072] Moreover, 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 telecommunications network 502 includes one or more Open-RAN (ORAN) network nodes. An ORAN network node is a network node in the telecommunications network 502 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 network nodes to implement one or more functionalities of any network node in the telecommunications network 502, including one or more access network nodes 510 and / or core network nodes 508.
[0073] Examples of an ORAN network node include an open radio unit (O-RU), an open distributed unit (O-DU), an open central unit (O-CU), including an O-CU control plane (O-CU-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). An ORAN network node may support aspecification by, for example, supporting an interface defined by the ORAN specification, such as an Al, Fl, Wl, El, E2, X2, Xn interface, an open fronthaul user plane interface, or an open fronthaul management plane interface. Moreover, an ORAN network 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.
[0074] The network nodes 510 facilitate direct or indirect connection of one or more UEs 512 to the core network 506 over one or more wireless connections. 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 500 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 500 may include and / or interface with any type of communication, telecommunication, data, cellular, radio network, and / or other similar type of system.
[0075] The UEs 512 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 510 and other communication devices. Similarly, the network nodes 508, 510 are arranged, capable, configured, and / or operable to communicate directly or indirectly (e.g., via other devices of telecommunications network 502) with the UEs 512 and / or with other network nodes or equipment in the telecommunications network 502 to enable and / or provide network access, such as wireless network access, and / or to perform other functions, such as administration in the telecommunications network 502. More specifically, UEs 512 may send messages, data, and / or other signals to network nodes 508, 510 or other elements of the telecommunications network 502 by transmitting such signals to the relevant device directly without the signals passing through any intervening devices or by transmitting such signals to the relevant device indirectly through an intervening device (or multiple intervening devices) that then transmit the signal to the relevant device. Similarly, network nodes 508, 510 may send messages, data, and other signals to UEs 5122, other network nodes 508, 510, and other devices in telecommunications network 502 directly or indirectly. As one specificexample, a core network node 108 may transmit a particular message to a UE 512 by transmitting the message to an access network node 510 that will then transmit the message to the intended UE 512. Similarly, a core network node 108 may receive a particular message from a UE 512 by receiving the message from an access network node 510 that itself received the message from the UE 512.
[0076] In the depicted example, the core network 506 connects elements of the access network 504 (e.g., one or more of the network nodes 510) to one or more host computing systems, such as host 516. 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 506 includes one or more core network nodes (e.g., core network node 508) of various types, one or more of which may be generally referred to as network nodes 508. Network nodes 508 are structured with hardware and software components. Features of these components may be substantially similar to those described with respect to the UEs, access network nodes, and / or hosts, such that the descriptions thereof are generally applicable to the corresponding components of the core network node 508. Example core network nodes provide functions of one or more of a Mobile Switching Center (MSC), Mobility Management Entity (MME), Home Subscriber Server (HSS), AMF, Session Management Function (SMF), AUSF, Subscription Identifier De-concealing function (SIDF), UDM, Security Edge Protection Proxy (SEPP), NEF, and / or a UPF.
[0077] The host 516 may be under the ownership or control of a service provider other than an operator or provider of the access network 504 and / or the telecommunications network 502. The host 516 may be operated by the service provider or on behalf of the service provider. The host 516 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.
[0078] As a whole, the communication system 500 of Figure 5 enables connectivity between the UEs, network nodes, and hosts. In that sense, the communication system 500 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)702.11 standards (Wi-Fi); and / or any other appropriate wireless communication standard, such as the Worldwide Interoperability for Microwave Access (Wi-Max), Bluetooth, Z-Wave, Near Field Communication (NFC) ZigBee, Li-Fi, and / or any low-power wide-area network (LPWAN) standards such as LoRa and Sigfox. Moreover, the communication system 500 may be configured to support multiple different standards, protocols, or other rule sets, with individual components supporting all of the relevant rule sets or with different components or sub-systems within the communication system 500 supporting different standards, protocols, or rule sets.
[0079] As one example, in certain embodiments, access network 504 may contain some access network nodes 510 that support 3GPP radio access technologies (RAT), such as LTE or NR, while other access network nodes 510 support (or the same access network nodes 510 additionally support) non-3GPP RATs, such as Wi-Fi or a proprietary RAT. As another example, telecommunications network 502 may support multiple generations of related communication standards (e.g., 4G and 5G 3GPP communication standards) and, as a result, may include an access network 104 and / or a core network 106 that supports multiple different standard generations or may include multiple access networks 104 and / or multiple core networks 106 with individual networks 104, 106 supporting different standard generations.
[0080] Telecommunications network 502 may support network slicing to provide different logical networks to different devices that are connected to the telecommunications network 502. For example, the telecommunications network 502 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.
[0081] In some examples, one or more of the UEs 512 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 504 on a predetermined schedule, when triggered by an internal or external event, or in response to requests from the access network 504. 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).
[0082] In the example, the hub 514 communicates with the access network 504 to facilitate indirect communication between one or more UEs (e.g., UE 512C and / or 512D) and network nodes (e.g., network node 510B). In some examples, the hub 514 may be a controller, router, contentsource and analytics, or any of the other communication devices described herein regarding UEs. For example, the hub 514 may be a broadband router enabling access to the core network 506 for the UEs. As another example, the hub 514 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 510, or by executable code, script, process, or other instructions in the hub 514.
[0083] As another example, the hub 514 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 514 may be a content source. For example, for a UE that is a VR headset, display, loudspeaker or other media delivery device, the hub 514 may retrieve VR assets, video, audio, or other media or data related to sensory information via a network node, which the hub 514 then provides to the UE either directly, after performing local processing, and / or after adding additional local content. In still another example, the hub 514 acts as a proxy server or orchestrator for the UEs, in particular if one or more of the UEs are low energy loT devices.
[0084] The hub 514 may have a constant / persistent or intermittent connection to the network node 510B. The hub 514 may also allow for a different communication scheme and / or schedule between the hub 514 and UEs (e.g., UE 512C and / or 512D), and between the hub 514 and the core network 506. In other examples, the hub 514 is connected to the core network 506 and / or one or more UEs via a wired connection. Moreover, the hub 514 may be configured to connect to an M2M service provider over the access network 504 and / or to another UE over a direct connection. In some scenarios, UEs may establish a wireless connection with the network nodes 510 while still connected via the hub 514 via a wired or wireless connection. In some embodiments, the hub 514 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 510B. In other embodiments, the hub 514 may be a nondedicated hub - that is, a device which is capable of operating to route communications between the UEs and network node 510B, but which is additionally capable of operating as a communication start and / or end point for certain data channels.
[0085] MMF 110 can be implemented in various ways in a mobile network (also referred to herein as a telecommunication network). In some embodiments, MMF 110 is part of the core network, and the various components of MMF 110 can be spread across different network functions (NFs). Figures 6A and 6B are block diagrams illustrating two non-limiting examples of implementation of MMF 110. Figure 6A shows MANO function 202 informing a Analytics Data Repository Function (ADRF) 600, which includes node topology 116 and node profile 112, on (1) a topology update (DataID=Topology Update) that includes a node id and information onneighboring nodes (“neighbors”) via a network analytics data function (NADF) data management storge request (Nadf DaataManagement StorageRequest); and (2) a profile update (DataID=Profile_Update) regarding capabilities of a node(s) that can host a ML model(s). The profile update in this example includes a node id, a load, a capacity, a location, and mobility of the node, as well as a timestamp. Alternatively, in some examples, instead of MANO 202, an operations support system (OSS) of the mobile network reports this information in cases where the nodes that host the ML models are 3GPP nodes, e.g., gNBs.
[0086] The example in Figure 6B includes exposure of a computer implemented method, as discussed herein with respect to certain embodiments, to an external third party(s). In this example, the external third party is an application function (AF) 602. AF 602, in this example, submits tasks that are executed in different parts of a core network depending on the requirements. The example in Figure 6B shows information flow between different 3 GPP nodes, specifically among network exposure function (NEF) 604 which includes endpoint 200; analytics logical function (ANLF) 606 which includes task clusters 114; MANO 202; model training logical function (MTLF) 608; and analytical data repository function (ADRF) 610 which includes node topology 116.
[0087] Referring to Figure 6B, in this non-limiting example, the flow starts with operation 1 in which AF 602 submits a task (Nnef EventExposure Subscription including EventID =Task Submission and task description), via NEF 604, to a network data and analytics function (NWDAF), specifically to ANLF 606. In operation 2, NEF 604 submits the task to ANLF 606 via Namf AnalyticsInfo Subscribe, AnalyticsID = Execute Task, which includes the task description. In operations 3-5, ANLF 606 uses MTLF 608 to cluster the task and select a proper ML model for the task, with the help of ADRF 610 stored information (e.g., the information discussed with respect to Figure 6A). In operation 6, once an action is selected, ANLF 606 directly contacts MANO 202 with ModelProvisionRequestAction to apply the action.
[0088] Figure 7 shows a node 700 in accordance with some embodiments. As used herein, a node refers to equipment capable, configured, arranged and / or operable to perform operations in accordance with respective embodiments.
[0089] In accordance with respective embodiments, node 700 may be configured to operate in communication system 500 of Figure 5, like access network nodes 510 or core network nodes 508. Examples of nodes include, but are not limited to, base stations (BSs) (e.g., radio base stations, Node Bs, evolved Node Bs (eNBs) and NR NodeBs (gNBs)), 0-RAN nodes or components of an 0-RAN node (e g., 0-RU, 0-DU, O-CU).
[0090] Nodes 700 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. Node 700 may be a relay node or a relay donor node controlling a relay. Nodes 700 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).
[0091] Other examples of nodes 700 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-cell / multicast coordination entities (MCEs), Operation and Maintenance (O&M) nodes, Operations Support System (OSS) nodes, SelfOrganizing Network (SON) nodes, positioning nodes (e.g., Evolved Serving Mobile Location Centers (E-SMLCs)), and / or Minimization of Drive Tests (MDTs).
[0092] In particular embodiments, node 700 includes a processing circuitry 702, a memory 704, a communication interface 706, and a power source 708. In general, in a particular embodiment of node 700, processing circuitry 702, memory 704, communication interface 706, and power source 708 may, in whole or in part, represent or include physical components common to or shared by one or more of the other elements of node 700.
[0093] The node 700 may be composed of multiple distinct network entities (e.g., a NodeB entity and a RNC entity, or a BTS entity and a BSC entity, etc.), which may each have or utilize their own respective physical components. In certain scenarios in which the node 700 comprises multiple such entities (e.g., BTS and BSC), one or more of the separate entities 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 node. In some embodiments, the node 700 may be configured to support multiple radio access technologies (RATs). In such embodiments, some components may be duplicated (e.g., separate memories 704 or portions of memory 704 for different RATs) and some components may be reused (e.g., a same antenna 710 may be shared by different RATs). The node 700 may also include multiple sets of the various illustrated components for different wireless technologies integrated into node 700, for example GSM, WCDMA, LTE, NR, Wi-Fi (e.g., according to an IEEE 802.11 familystandard), 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 node 700.
[0094] The processing circuitry 702 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 components, such as the memory 704 and / or logical functions 112, 114, 116, and / or 118, to provide node 700 functionality.
[0095] In some embodiments, the processing circuitry 702 includes a system on a chip (SOC). In some embodiments, the processing circuitry 702 includes one or more of radio frequency (RF) transceiver circuitry 712 and baseband processing circuitry 714. In some embodiments, the RF transceiver circuitry 712 and the baseband processing circuitry 714 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 712 and baseband processing circuitry 714 may be on the same chip or set of chips, boards, or units.
[0096] The memory 704 and / or logical functions 112, 114, 116, and / or 118 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 computerexecutable memory devices that store information, data, and / or instructions that may be used by the processing circuitry 702. The memory 704 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 702 and utilized by the node 700. The memory 704 and / or logical functions 112, 114, 116, and / or 118 may be used to store any calculations made by the processing circuitry 702 and / or any data received via the communication interface 706. In some embodiments, the processing circuitry 702, memory 704, and / or logical functions 112, 114, 116, and / or 118 is integrated.
[0097] The communication interface 706 is used in wired or wireless communication of signaling and / or data with UEs, other network nodes, and / or any other network equipment. In the illustrated embodiment, communication interface 706 comprises port(s) / terminal(s) 716 to send andreceive data, for example to and from a network over a wired connection. In particular embodiments, node 700 may be capable of wireless communication and communication interface 706 may also include radio front-end circuitry 718 that may be coupled to, or in certain embodiments a part of, an antenna 710. Particular embodiments of radio front-end circuitry 718 include filter(s) 720 and amplifier(s) 722. The radio front-end circuitry 718 may be connected to an antenna 710 and processing circuitry 702. The radio front-end circuitry may be configured to condition signals communicated between antenna 710 and processing circuitry 702. The radio frontend circuitry 718 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 718 may convert the digital data into a radio signal(s) having the appropriate channel and bandwidth parameters using a combination of filters 720 and / or amplifiers 722. The radio signal(s) may then be transmitted via the antenna 710.Similarly, when receiving data, the antenna 710 may collect radio signals which are then converted into digital data by the radio front-end circuitry 718. The digital data may be passed to the processing circuitry 702. In other embodiments, the communication interface may comprise different components and / or different combinations of components.
[0098] In certain alternative embodiments, node 700 may be capable of wireless communication but does not include separate radio front-end circuitry 718, instead, the processing circuitry 702 includes radio front-end circuitry and is connected to the antenna 710. Similarly, in some embodiments, all or some of the RF transceiver circuitry 712 is part of the communication interface 706. In still other embodiments, the communication interface 706 includes one or more ports or terminals 716, the radio front-end circuitry 718, and the RF transceiver circuitry 712, as part of a radio unit (not shown), and the communication interface 706 communicates with the baseband processing circuitry 714, which is part of a digital unit (not shown).
[0099] The antenna 710 may include one or more antennas, or antenna arrays, configured to send and / or receive wireless signals. The antenna 710 may be coupled to the radio front-end circuitry 718 and may be any type of antenna capable of transmitting and receiving data and / or signals wirelessly. In certain embodiments, the antenna 710 is separate from the node 700 and connectable to the node 700 through one or more interfaces or ports.
[0100] The antenna 710, communication interface 706, and / or the processing circuitry 702 may be configured to perform some or all of the receiving operations and / or obtaining operations described herein as being performed by the node 700. Any information, data and / or signals may be received from a UE, another network node and / or any other network equipment. Similarly, the antenna 710, the communication interface 706, and / or the processing circuitry 702 may beconfigured to perform some or all of the transmitting or sending operations described herein as being performed by the node 700. Any information, data and / or signals may be transmitted to a UE, another network node and / or any other network equipment.
[0101] The power source 708 provides power to the various components of node 700 in a form suitable for the respective components (e.g., at a voltage and current level needed for each respective component). The power source 708 may further comprise, or be coupled to, power management circuitry to supply the components of the node 700 with power for performing the functionality described herein. For example, the node 700 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 708. As a further example, the power source 708 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.
[0102] Embodiments of the node 700 may include additional components beyond those shown in Figure 7 for providing certain aspects of the 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 node 700 may include user interface equipment to allow input of information into the node 700 and to allow output of information from the node 700. This may allow a user to perform diagnostic, maintenance, repair, and other administrative functions for the node 700.
[0103] Figure 8 is a block diagram illustrating a virtualization environment 800 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 800 hosted by one or more of hardware nodes, such as a hardware computing device that operates as an access network node, UE, core network node, or host. Further, in embodiments in which a 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 800includes 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.
[0104] Applications 802 (which may alternatively be called software instances, virtual appliances, network functions, virtual nodes, virtual network functions, etc.) are run in the virtualization environment 800 to implement some of the features, functions, and / or benefits of some of the embodiments disclosed herein.
[0105] Hardware 804 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 806 (also referred to as hypervisors or virtual machine monitors (VMMs)), provide VM 808A and VM 808B (which may be collectively referred to as VMs 808), and / or perform any of the functions, features and / or benefits described in relation with some embodiments described herein. The virtualization layer 806 may present a virtual operating platform that appears like networking hardware to one or more of the VMs 808.
[0106] The VMs 808 comprise virtual processing, virtual memory, virtual networking or interface and virtual storage, and may be run by virtualization layer 806. Different embodiments of the instance of a virtual appliance 802 may be implemented on one or more of VMs 808, 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.
[0107] In the context of NFV, each of the VMs 808 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 808, and that part of hardware 804 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 of the VMs 808 on top of the hardware 804 and corresponds to an application 802.
[0108] Hardware 804 may be implemented in a standalone network node with generic or specific components. Hardware 804 may implement some functions via virtualization.Alternatively, hardware 804 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 andorchestration 810, which, among others, oversees lifecycle management of applications 802. In some embodiments, hardware 804 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 812 which may alternatively be used for communication between hardware nodes and radio units.
[0109] In certain embodiments, a node 110, 508, 700, 802 is configured for ML model management orchestration in a telecommunication network. The node includes a logical function 112, 114, 116, 118, and / or 810; processing circuitry 702; and at least one memory 704 connected to the processing circuitry and storing instructions that when executed by the processing circuitry causes the node to perform operations. The operations include to perform some or all of the functionality described herein.
[0110] Although nodes described herein may include the illustrated combination of hardware components, other embodiments may comprise nodes with different combinations of components. It is to be understood that these nodes 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 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.
[0111] Moreover, while components are depicted as single boxes located within a larger box, or nested within multiple boxes, in practice, nodes 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.
[0112] 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 certainembodiments 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 node, but are enjoyed by the node as a whole, and / or by end users and a wireless network generally.
[0113] In the above-description of certain embodiments of the present disclosure, it is to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the present disclosure. Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which concepts of the present disclosure belong. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of this specification and the relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
[0114] When an element is referred to as being “connected”, “coupled”, “responsive”, or variants thereof to another element, it can be directly connected, coupled, or responsive to the other element or intervening elements may be present. In contrast, when an element is referred to as being “directly connected”, “directly coupled”, “directly responsive”, or variants thereof to another element, there are no intervening elements present. Like numbers refer to like elements throughout. Furthermore, “coupled”, “connected”, “responsive”, or variants thereof as used herein may include wirelessly coupled, connected, or responsive. As used herein, the singular forms “a”, “an” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. Well-known functions or constructions may not be described in detail for brevity and / or clarity. The term “and / or” (abbreviated “ / ”) includes any and all combinations of one or more of the associated listed items.
[0115] It will be understood that although the terms first, second, third, etc. may be used herein to describe various elements / operations, these elements / operations should not be limited by these terms. These terms are only used to distinguish one element / operation from another element / operation. Thus a first element / operation in some embodiments could be termed a secondelement / operation in other embodiments without departing from the teachings of concepts of the present disclosure. The same reference numerals or the same reference designators denote the same or similar elements throughout the specification.
[0116] As used herein, the terms “comprise”, “comprising”, “comprises”, “include”, “including”, “includes”, “have”, “has”, “having”, or variants thereof are open-ended, and include one or more stated features, integers, elements, steps, components, or functions but does not preclude the presence or addition of one or more other features, integers, elements, steps, components, functions, or groups thereof. Furthermore, as used herein, the common abbreviation “e.g ”, which derives from the Latin phrase “exempli gratia,” may be used to introduce or specify a general example or examples of a previously mentioned item, and is not intended to be limiting of such item. The common abbreviation “i.e ”, which derives from the Latin phrase “id est,” may be used to specify a particular item from a more general recitation.
[0117] Example embodiments are described herein with reference to block diagrams and / or flowchart illustrations of computer-implemented methods, apparatus (systems and / or devices) and / or computer program products. It is understood that a block of the block diagrams and / or flowchart illustrations, and combinations of blocks in the block diagrams and / or flowchart illustrations, can be implemented by computer program instructions that are performed by one or more computer circuits. These computer program instructions may be provided to a processor circuit of a general purpose computer circuit, special purpose computer circuit, and / or other programmable data processing circuit to produce a machine, such that the instructions, which execute via the processor of the computer and / or other programmable data processing apparatus, transform and control transistors, values stored in memory locations, and other hardware components within such circuitry to implement the functions / acts specified in the block diagrams and / or flowchart block or blocks, and thereby create means (functionality) and / or structure for implementing the functions / acts specified in the block diagrams and / or flowchart block(s).
[0118] These computer program instructions may also be stored in a tangible computer-readable medium that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable medium produce an article of manufacture including instructions which implement the functions / acts specified in the block diagrams and / or flowchart block or blocks. Accordingly, embodiments of the present disclosure may be embodied in hardware and / or in software (including firmware, resident software, micro-code, etc.) that runs on a processor such as a digital signal processor, which may collectively be referred to as “circuitry,” “a module” or variants thereof.
[0119] It should also be noted that in some alternate implementations, the functions / acts noted in the blocks may occur out of the order noted in the flowcharts. For example, two blocks shown in succession may in fact be executed substantially concurrently or the blocks may sometimes be executed in the reverse order, depending upon the functionality / acts involved. Moreover, the functionality of a given block of the flowcharts and / or block diagrams may be separated into multiple blocks and / or the functionality of two or more blocks of the flowcharts and / or block diagrams may be at least partially integrated. Finally, other blocks may be added / inserted between the blocks that are illustrated, and / or blocks / operations may be omitted without departing from the scope of the present disclosure. Moreover, although some of the diagrams include arrows on communication paths to show a primary direction of communication, it is to be understood that communication may occur in the opposite direction to the depicted arrows.
[0120] Many variations and modifications can be made to the embodiments without substantially departing from the principles of the present disclosure. All such variations and modifications are intended to be included herein within the scope of present disclosure.Accordingly, the above disclosed subject matter is to be considered illustrative, and not restrictive, and the examples of embodiments are intended to cover all such modifications, enhancements, and other embodiments, which fall within the spirit and scope of the present disclosure. Thus, to the maximum extent allowed by law, the scope of the present disclosure is to be determined by the broadest permissible interpretation of the present disclosure including the examples of embodiments and their equivalents, and shall not be restricted or limited by the foregoing detailed description.
Claims
Claims:
1. A computer-implemented method for machine learning, ML, model management orchestration in a telecommunication network, the method comprising:accessing (408) at least one task directed towards at least one ML model in a plurality of ML models in the telecommunications network;classifying ( 10) the at least one task into one or more domains in a plurality of domains in the telecommunications network, the classifying based on a classification of the at least one task into one or more domains to which the at least one task belongs; androuting (412) the at least one task to a ML model in a domain based on the classification.
2. The computer-implemented method of Claim 1, further comprising:accessing (400) node information from a plurality of nodes associated with the telecommunications network, the node information comprising respective availability of respective nodes in the plurality of nodes, respective capabilities of the respective nodes, and at least one respective task directed by the respective nodes towards the telecommunication network; and creating (402) a respective profile of a respective node and a respective task description of the at least one respective task.
3. The computer-implemented method of Claim 2, wherein a respective profile comprises at least an identifier of a respective node, a capacity of the respective node, and a load of the respective node.
4. The computer-implemented method of any one of Claims 2 to 3, further comprising: creating (404) a description of a relationship among the plurality of nodes.
5. The computer-implemented method of Claim 4, wherein the description of the relationship among the plurality of nodes comprises, per node, an identifier of a respective node, a location of the respective node, and neighboring relationships between nodes.
6. The computer-implemented method of Claim 5, further comprising:constructing (406) a node topology based on the description of the relationship among the plurality of nodes.
267. The computer-implemented method of any one of Claims 1 to 6, wherein the classifying (410) comprises clustering of tasks per domain, and the classification of the at least task belongs to one cluster comprising one domain.
8. The computer-implemented method of any one of Claims 1 to 6, wherein the classifying (410) comprises clustering of tasks per domain, the classification of the at least one task belongs to a plurality of clusters, and respective clusters in the plurality of clusters respectively comprise a different domain.
9. The computer-implemented method of any one of Claims 2 to 8, wherein the classifying (410) at least one respective task into one or more domains is based on using the node information to cluster the at least one respective task by domain.
10. The computer-implemented method of any one of Claims 2 to 9, wherein the classifying (410) at least one respective task comprises clustering at least one respective task into one or more clusters based on the at least one respective task description of the at least one respective task.
11. The computer-implemented method of Claim 10, wherein the classifying (410) further comprises clustering at least one respective task by area of origin in the telecommunications network of the at least respective task.
12. The computer-implemented method of any one of Claims 2 to 11, wherein the clustering at least one respective task comprisesconverting the at least one respective task description of the at least one respective task into a vector representation, andassigning the at least one respective task to one or more clusters based on a similarity of the vector representation to a cluster or a centroid of a cluster.
13. The computer-implemented method of any one of Claims 2 to 12, further comprising: repeating (414) classifying of at least one task into one or more domains based on at least one of (i) accessing a new task description, after a last classifying, that exceed a threshold value, and (ii) an elapsed time period.
14. The computer-implemented method of any one of Claims 1 to 13, further comprising: initiating ( 16) at least one orchestration action for at least one ML model based on the classification of the at least one task into one or more domains.
15. The computer-implemented method of Claim 14, wherein initiating (416) at least one orchestration action for the at least one ML model is further based on available resources of the one or more domains for the classified at least one task.
16. The computer-implemented method of any one of Claims 14 to 15, wherein initiating (416) at least one orchestration action comprises:accessing information comprising the classification, information about a node from which the at least one task originated, and information about a relationship among a plurality of nodes associated with the telecommunications network; anddetermining one or more orchestration actions based on the accessed information.
17. The computer-implemented method of any one of Claims 14 to 16, wherein the at least one orchestration action comprises at least one of:creating a new ML model instance of a given ML model identification at an identified node, transferring a ML model instance from a first node to a second node,distilling a ML model instance to a simpler ML model instance based on use of a ML model optimization, andremoving a ML model instance of a given ML model identification at an identified node.
18. The computer-implemented method of any one of Claims 14 to 17, wherein initiating (416) the at least one orchestration action is triggered by at least one of an elapsed time period after accessing the at least one task, expiration of a time period, and an event in the telecommunication network.
19. The computer-implemented method of any one of Claims 1 to 18, wherein a domain comprises at least one of a radio access network, RAN, a transport network, a core network, and / or at least one function of the RAN, the transport network, and the core network.
20. The computer-implemented method of any one of Claims 1 to 19, further comprising:exposing (418) the computer-implemented method to an application function.
21. The computer-implemented method of any one of Claims 1 to 20, wherein the ML model comprises a generative model.
22. A node (110, 508, 700, 802) having a logical function (112, 114, 116, 118, 810), the node comprising:processing circuitry (702); andmemory (704) coupled with the processing circuitry, wherein the memory includes instructions that when executed by the processing circuitry causes the logical function (112, 114, 116, 118, 810) to perform operations comprising:access at least one task directed towards at least one ML model in a plurality of ML models in a telecommunications network;classify the at least one task into one or more domains in a plurality of domains in the telecommunications network, the classifying based on a classification of the at least one task into one or more domains to which the at least one task belongs; androute the at least one task to a ML model in a domain based on the classification.
23. The node (110, 508, 700, 802) of Claim 22 wherein the memory includes further instructions that when executed by the processing circuitry causes the logical function (112, 114, 116, 118, 810) to perform operations according to any one of Claims 2 to 21.
24. The node (110, 508, 700, 802) of any one of Claims 22 to 23, wherein the logical function is distributed in one or more additional nodes associated with the telecommunications network.
25. A computer program product comprising a non-transitory storage medium including program code to be executed by processing circuitry (702) of a node (110, 508, 700, 802) whereby execution of the program code causes a logical function (112, 114, 116, 118, 810) to perform operations comprising:access at least one task directed towards at least one ML model in a plurality of ML models in a telecommunications network;classify the at least one task into one or more domains in a plurality of domains in the29telecommunications network, the classifying based on a classification of the at least one task into one or more domains to which the at least one task belongs; androute the at least one task to a ML model in a domain based on the classification.
26. The computer program product of Claim 25, wherein the non-transitory storage medium includes further program code, whereby execution of the program code causes the logical function (112, 114, 116, 118, 810) to perform according to any one of Claims 2 to 21.
27. The computer program product of Claim 25, wherein the logical function is distributed in one or more additional nodes associated with the telecommunications network.30
