Apparatus and method for mapping machine learning-based applications to intent logic units

Abstract

A method includes sending a request for capability information, where the request is sent to at least one of an ML-based application, an ML application coordinator, and an ML capability producer; receiving capability information representing capabilities of one or more machine learning (ML)-based applications; generating one or more intent logic units (ILUs), where the ILUs are mapped with at least one corresponding ML-based application, where the ILUs are configured to initiate execution of the ML-based application corresponding to at least one of the ILU maps; one or more ILUs are stored in an intent logic library, wherein the ILUs are stored in association with capability information derived for the ILUs from among the capability information received for the ML-based application corresponding to at least one of the ILUs maps.

Description

Technical Field

[0001] Various example embodiments generally relate to apparatus, methods, and computer programs for mapping ML-based applications to intent logic units. Background Art

[0002] Due to the increasing complexity of communication networks, especially mobile communication networks, there is always a need for more automation and abstraction. One solution to abstraction is through the use of intent, the so-called intent-based networking (IBN). Therein, the network should be able to respond to the user's intent without the user (e.g. operator) specifying the technical details of how his expected results should be achieved through the underlying architecture.

[0003] Technical report 3GPP TR 28.812 V17.1.0 (dated 2020-12) (Technical Specification Group Services and System Aspects; Telecommunication Management; Scenario study of intent-driven management services for mobile networks) provides example use cases for intent-driven management for mobile networks.

[0004] The complexity of a specified intent can vary greatly. A simple intent can be implemented with a single command to a single network object. An intent can allow for solving a specific network problem or achieving a specific network goal.

[0005] To implement complex intent, it may be necessary to perform continuous network monitoring, data processing, and infer the best commands to be applied to single or multiple network objects. Summary of the invention

[0006] The protection scope is defined by the independent claims. Embodiments, examples and features described in this specification that do not belong to the protection scope (if any) should be interpreted as individual examples that help understand the various embodiments and examples that fall within the protection scope.

[0007] According to a first aspect, an apparatus is provided, comprising a component for executing a method, the method comprising: sending a request for capability information, wherein the request is sent to at least one of the following items: an ML-based application, an ML application coordinator, and an ML capability producer; receiving capability information representing capabilities of one or more machine learning (ML)-based applications; generating one or more intent logic units (ILUs), wherein the ILU is mapped to at least one corresponding ML-based application, wherein the ILU is configured to initiate execution of at least one corresponding ML-based application mapped to the ILU; storing the one or more ILUs in an intent logic library, wherein the ILU is stored in association with the capability information, and the capability information is derived for the ILU from the capability information received for at least one corresponding ML-based application mapped to the ILU.

[0008] The capability information received for the ML-based application may include: information representing a type of function performed by the ML-based application for at least one communication network, wherein the type is at least one of: at least one of an optimization function, a control function, an analysis function, and an ML coordination function.

[0009] The ILU may be configured to initiate execution of at least one corresponding ML-based application and / or initiate an ML coordination function configured to monitor at least one corresponding ML-based application.

[0010] The capability information received for the ML-based application may include information indicating one or more objects or one or more object types that perform the function.

[0011] The capability information received for the ML-based application may include information representing at least one configuration parameter of an object or object type that performs the function.

[0012] The capability information received for the ML-based application may include information representing goals and / or constraints for performing the function, wherein the goals and / or constraints are defined based on at least one network metric.

[0013] The capability information received for the ML-based application may include information representing configuration parameters of the ML model implemented by the ML-based application.

[0014] The method may include storing mapping information in an intent logic library for mapping performed between the ILU and at least one corresponding ML-based application.

[0015] The method may include: receiving a request including intent information representing an intent to be implemented for at least one communication network; identifying one or more ILUs configured to initiate execution of one or more corresponding ML-based applications based on capability information stored in an intent logic library, the one or more corresponding ML-based applications being suitable for implementing the intent for at least one communication network; and executing the identified one or more ILUs.

[0016] The method may include configuring one or more ML-based applications mapped with the identified one or more ILUs.

[0017] According to a second aspect, a method is provided, comprising: sending a request for capability information, wherein the request is sent to at least one of: an ML-based application, an ML application coordinator, and an ML capability producer; receiving capability information representing capabilities of one or more machine learning (ML)-based applications; generating one or more intent logic units (ILUs), wherein the ILU is mapped to at least one corresponding ML-based application, wherein the ILU is configured to initiate execution of at least one corresponding ML-based application mapped to the ILU; storing the one or more ILUs in an intent logic library, wherein the ILU is stored in association with the capability information, wherein the capability information is derived for the ILU from the capability information received for at least one corresponding ML-based application mapped to the ILU.

[0018] Typically, the apparatus according to the first aspect comprises a component for performing one or more or all steps of the method according to the second aspect. The component may include a circuit system configured to perform one or more or all steps of the method. The component may include at least one processor and at least one memory including computer program code, the at least one memory and the computer program code being configured to, together with the at least one processor, cause the apparatus to perform one or more or all steps of the method.

[0019] According to another aspect, a non-transitory computer-readable medium is provided, the medium including program instructions stored thereon for performing at least the following operations: sending a request for capability information, wherein the request is sent to at least one of the following items: an ML-based application, an ML application coordinator, and an ML capability producer; receiving capability information representing the capabilities of one or more machine learning (ML)-based applications; generating one or more intent logic units (ILUs), wherein the ILU is mapped with at least one corresponding ML-based application, wherein the ILU is configured to initiate execution of at least one corresponding ML-based application mapped with the ILU; storing one or more ILUs in an intent logic library, wherein the ILU is stored in association with capability information derived for the ILU from capability information received for at least one corresponding ML-based application mapped with the ILU. In general, a non-transitory computer-readable medium may include program instructions stored thereon for performing one or more or all steps of the method according to the second aspect.

[0020] According to a third aspect, a device is provided, comprising a component for executing a method, the method comprising: receiving a request for capability information from an intent logic execution platform; in response to the request, sending capability information representing the capabilities of one or more machine learning (ML)-based applications; executing one or more machine learning (ML)-based applications to implement an intent, wherein the execution is initiated by at least one intent logic unit mapped to the one or more ML-based applications.

[0021] According to a fourth aspect, a method is provided, comprising: receiving a request for capability information from an intent logic execution platform; sending capability information representing capabilities of one or more machine learning (ML)-based applications in response to the request; executing one or more machine learning (ML)-based applications to implement an intent, wherein the execution is initiated by at least one intent logic unit mapped to the one or more ML-based applications.

[0022] Typically, the apparatus according to the third aspect comprises a component for performing one or more or all steps of the method according to the fourth aspect. The component may include a circuit system configured to perform one or more or all steps of the method. The component may include at least one processor and at least one memory including computer program code, the at least one memory and the computer program code being configured to, together with the at least one processor, cause the apparatus to perform one or more or all steps of the method.

[0023] According to another aspect, a non-transitory computer-readable medium is provided, the medium including program instructions stored thereon for performing at least the following operations: receiving a request for capability information from an intent logic execution platform; in response to the request, sending capability information representing the capabilities of one or more machine learning (ML)-based applications; executing one or more machine learning (ML)-based applications to implement the intent, wherein the execution is initiated by at least one intent logic unit mapped to the one or more ML-based applications. In general, a non-transitory computer-readable medium may include program instructions stored thereon for performing one or more or all steps of the method according to the fourth aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate example embodiments together with the general description given above and the detailed description given below.

[0025] Figure 1 is a schematic diagram illustrating an exemplary system for implementing the intent according to an example.

[0026] Figure 2 is a schematic diagram illustrating an exemplary system for implementing the intent according to an example.

[0027] Figure 3 is a schematic diagram illustrating aspects of the relationship between an ML application, an intent logic execution platform, and an intent logic unit according to an example.

[0028] Figure 4 is a schematic diagram illustrating aspects of the relationship between an ML application, an intent logic execution platform, and an intent logic unit according to an example.

[0029] Figure 5 is a flow chart illustrating a method implemented according to the intent of the examples.

[0030] Figure 6 is a flow chart illustrating a method implemented according to the intent of the examples.

[0031] Figure 7 is a flow chart illustrating a method implemented according to the intent of the examples.

[0032] Figure 8 is a flow chart illustrating a method implemented according to the intent of the examples.

[0033] Fig. 9 is an exemplary hardware structure of an apparatus according to an example.

[0034] It should be noted that these figures are intended to illustrate various aspects of the apparatus, methods, and structures used in the example embodiments described herein. The use of similar or identical reference numerals in the various figures is intended to indicate the presence of similar or identical elements or features. DETAILED DESCRIPTION

[0035] Detailed example embodiments are disclosed herein. However, the specific structural and functional details disclosed herein are only meant to describe the example embodiments. Therefore, these embodiments are shown in the accompanying drawings in the form of illustrative examples and will be described in detail herein to provide a thorough understanding of various aspects. However, it will be understood by those of ordinary skill in the art that the example embodiments are capable of various modifications and alternative forms and can be practiced without all the specific details. In addition, systems and processes may be shown in block diagrams so as not to obscure the example embodiments in unnecessary details. In other examples, well-known processes, structures, and techniques may be shown without unnecessary details to avoid obscuring the example embodiments.

[0036] The exemplary embodiments provide apparatus, methods, systems and computer programs for mapping machine learning (ML) based applications to intent logic units. The exemplary embodiments provide apparatus, methods, systems and computer programs for intent implementation.

[0037] The term "ML functionality" is used herein to refer to an ML-based application or an ML orchestrator. The term "ML-based application" or "ML application" is used herein to refer to a software application that uses AI (artificial intelligence) / ML techniques, for example by implementing one or more ML models. The term "ML coordination functionality" or "ML orchestrator" is used herein to refer to a software application that is configured to coordinate the execution of an ML application. Coordination of the ML application may include performing at least one of: synchronizing, triggering, configuring, monitoring, and controlling the execution of the coordinated ML application.

[0038] Exemplary applications of ML include, but are not limited to: speech recognition; image processing / computer vision; natural language processing; information retrieval; personalization and recommendations; robotics, data analytics (including predictive and prescriptive analytics); use cases in the design and / or planning and / or optimization and / or configuration and / or control and / or management of communication systems and / or networks.

[0039] Exemplary use cases may include, but are not limited to:

[0040] - Use cases related to the physical layer of communication networks, such as modulation, coding, decoding, signal detection, channel estimation, prediction, compression, interference mitigation;

[0041] - Use cases related to the medium access control layer of communication networks, such as multiple access and resource allocation (e.g., power control, scheduling, spectrum management);

[0042] -Channel modeling;

[0043] - Network optimization;

[0044] -Cell capacity estimation in cellular networks;

[0045] -routing;

[0046] - Resource management;

[0047] -Data business management;

[0048] - Security and anomaly detection;

[0049] -Root cause analysis;

[0050] -Transmission protocol design and optimization;

[0051] - User / network / application behavior analysis / prediction;

[0052] - transport layer congestion control;

[0053] -User experience modeling and optimization;

[0054] -User mobility and location management;

[0055] - Network slicing, network virtualization and software-defined networking;

[0056] - Nonlinear impairment compensation in optical networks (e.g., visible light communication, fiber-optic communication, and fiber-wireless converged networks), and

[0057] - Transmission quality estimation and optical performance monitoring in optical networks.

[0058] Exemplary analysis and / or decision-making functions include, but are not limited to: coverage analysis, coverage problem analysis, switching problem analysis, fault detection, interference detection, coverage optimization, capacity optimization, switching optimization, interference reduction, energy saving optimization,

[0059] In the context of this document, an intent may correspond to a goal to be met to achieve a target (e.g., target state, target location, target configuration, target quality, etc.) for one or more devices / entities interacting with and / or within a communication network. An intent may, for example, define a network management goal or a network deployment goal. An intent does not define the necessary steps to achieve a goal. The expressed intent may need to be translated into executable tasks to achieve the goal.

[0060] Intents can vary in complexity, from simple intents that can be implemented with a single command to a specific network object, to very complex intents that may require continuous network monitoring and inference of the best approach, resulting in different commands and involving multiple network nodes through the use of ML-based solutions.

[0061] The purpose of IBN (Intent-Based Networking) is to be able to respond to operator intent without specifying the technical details of how the underlying architecture implements the intent. Example intents may include, but are not limited to:

[0062] - Collect / obtain carrier aggregation statistics for all cells in city X;

[0063] - Restrict / deny high mobility users from switching to small cells;

[0064] - Allows load balancing to cell Y or small cells or only urban cells;

[0065] -Relocate the base station from controller A to controller B.

[0066] Generally speaking, if Figure 1 As shown in the system of , a user may access an intent specification platform (ISP) 120 through which a user (e.g., a human, an operator of an associated communication network, or a user device) specifies an intent to be achieved. Several types of user interfaces 110 may be used to specify intents, including, for example: a graphical user interface (GUI), a command line interface (CLI), a text interface (e.g., a text file), an audio interface that captures user audio commands, etc.

[0067] ISP 120 may be configured to analyze input data expressing an intent. ISP 120 may include mechanisms for enforcing grammatical rules and checking the correctness and completeness of a specified intent.

[0068] ISP 120, for example, can be configured to analyze (e.g., parse) input data expressing an intent provided by a user in order to identify data fields that conform to a predefined intent specification syntax. Several data processing methods can be used to analyze a specified intent and decompose the intent into its constituent sub-parts to facilitate identification of tasks / sub-tasks to be performed to achieve the intent.

[0069] Patent application published under number WO2021 / 164878A1 (which is incorporated herein by reference) discloses an example ISP that is configured to receive a request from a user through a user interface, wherein the request uses a specific intent specification syntax and may include a control object of a communication network and a verb indicating what action must be performed for the control object.

[0070] The intent execution interface 125 is used to exchange information between the ISP and the intent realization system 130. The intent execution interface 125 can be used, for example, to send a given intent to the intent realization system 130 for execution, or to receive a report on the success of the execution from the intent realization system 130. The intent realization system 130 itself is connected to the network functions and / or processes 140, which perform tasks to realize the received intent.

[0071] Depend on Figure 2 A system 200 for implementing user intent using an intent logic unit (ILU) is illustrated. The system 200 may be part of a communication network. Exemplary communication networks include, but are not limited to, a radio communication network, a wired or wireless communication network, an optical fiber-based communication network, an optical wireless communication network, a satellite communication network, or any combination thereof.

[0072] System 200 includes intent specification platform (ISP) 120 and intent fulfillment system (IFS) 230. Figure 1 Similar to the system of , a user may access an intent specification platform (ISP) 120 through a user interface 110 to allow the user to specify intents.

[0073] and Figure 1 Similar to the system of , ISP 120 communicates with IFS 230 through intention execution interface 125. Intent execution interface 125 is used to exchange information between ISP 120 and IFS 230. Intent execution interface 125 can be used to, for example, send execution intentions to IFS 230 or receive reports on execution success from IFS 230.

[0074] In this embodiment, the IFS 230 is implemented as an Intent Logic Execution Platform (ILEP) 230 and is configured to execute one or more Intent Logic Units (ILUs). The ILUs may be stored in and retrieved from the Intent Logic Library (ILL) 135 by the ILEP 230. The ILEP 230 itself is connected to the network functions and processes 140 that allow the implementation of intents. Intent execution is achieved via the Intent Logic Execution Platform (ILEP), which identifies the appropriate ILU and executes the ILU onto the relevant communication network or its network objects.

[0075] The ILU includes instructions (e.g., logic or command sequences) to be executed. The ILU includes instructions that, when executed by a processor, cause a host device or system (e.g., an Intent Implementation System (IFS) or Intent Logic Execution Platform (ILEP)) to initiate execution of one or more corresponding tasks. The instructions may be in any language (any computer programming language or scripting language, including but not limited to assembly language, C, C++, Basic, SQL, MySQL, HTML, PHP, Python, Java, Javascript, etc.) and in any format (script, binary code, executable code, precompiled code, interpretable code, etc.).

[0076] An ILU can be viewed as a wrapper around a sequence of logic or commands that may need to be executed to achieve a corresponding intent. An ILU may correspond to a software unit configured to initiate and / or configure the execution of one or more tasks, which may be, for example, any tasks or jobs related to network operations. These tasks may be executed to achieve a corresponding intent. Multiple ILUs may be combined to form another ILU.

[0077] An executable task may be any function and / or process or combination thereof that can be performed by a device. Functions and / or processes may be performed using one or more software units and one or more hardware units in any combination.

[0078] Executable tasks can be, but are not limited to:

[0079] - Network management tasks, including, for example, QoS management tasks, making handover decisions, load balancing tasks, configuring network devices or network cells, and implementing network access control;

[0080] -Identify, develop, activate and implement network management strategies.

[0081] To implement complex intents, it may be desirable or necessary to use a single or multiple ML models (executed in parallel or sequentially). The approach to using ML models to implement a given intent may depend on, for example, the complexity of the intent, the availability of ML models, and their suitability for implementing the given intent.

[0082] The present disclosure relates to using intents as an abstraction means, and automatically implementing them through machine learning (ML) based solutions mapped to one or more intent logical units (ILUs). Specified intents (e.g., intents specified by an operator of the communication network) can be implemented using ML functions available in the communication network. Specified intents can be implemented in an automated manner by using ML functions mapped to the ILUs.

[0083] In order to enable the use of ML techniques to realize intents, a capability discovery and disclosure mechanism for ML-based applications available in a communication network is provided. Once the relevant ML-based applications are deployed, the disclosed capabilities can be available.

[0084] In order to allow the use of ML-based applications to achieve specified intents (expressed by users (e.g. network operators)), it is proposed to exploit the public capabilities of ML-based applications in order to derive new ILUs that map to one or more ML-based applications or a combination of ML-based applications coordinated by an ML coordinator.

[0085] Using the capability discovery and disclosure mechanism allows the intent fulfillment system to determine the correct ML-based application to be used to fulfill a given intent. In addition, the intent fulfillment system can link ILUs and / or ML-based applications to satisfy a given intent.

[0086] By using ML-based applications, there can be different combinations / approaches to achieve intent:

[0087] There may be a one-to-one or one-to-many mapping between ILU and ML-based applications;

[0088] - To achieve a given intent, different combinations of ILU and ML-based applications can be used;

[0089] - Complex intents can be decomposed into intent components, where each intent component can be implemented by using one or more ILUs and mapping with one or more ML-based applications (in parallel or sequential execution).

[0090] Each ILU can be mapped to one or more corresponding ML-based applications from which the ILU is derived. The capabilities of the ILU correspond to the capabilities of the one or more ML-based applications mapped to the ILU. Then, using the capabilities of the new ILU, the newly generated ILU can be used to satisfy the intent (e.g., the intent specified by the operator for the communication network). In this way, the specified intent can be mapped to an ML-based solution via the ILU where applicable and to the extent possible.

[0091] Furthermore, a given intent may be mapped to other ILUs that are not themselves mapped to ML-based applications. For example, a non-ML-based solution (logic or commands) may also be required to fully implement the intent. In this specification, for simplicity reasons, we consider example intents or parts of intents that can be achieved through ML-based solutions, while non-ML-based commands may still be applicable to fully implement a given intent.

[0092] The ILEP may generate one or more intent logical units (ILUs), wherein there is a mapping between the one or more ILUs and the one or more ML-based applications. The ILU may be mapped with one or more corresponding ML-based applications, and the ILU is configured to initiate execution of the one or more corresponding ML-based applications mapped with the ILU.

[0093] Each ILU mapped to one or more ML applications is configured to initiate execution of the one or more ML applications directly or through an ML coordination function. Initiating execution of the ML application may include, for example, sending configuration parameters to the initiated ML application through the ILEP and / or the mapped ILU. The configuration parameters may be sent before and / or at and / or after triggering execution of the initiated ML application. Initiation may be performed using a remote procedure call, a web interface, or any suitable software technique to trigger execution of the ML application that may be executed on a remote device.

[0094] Although the ILU is executed by the ILEP, each mapped ML application may be executed by a host device (e.g., a network entity) that is physically distinct and separate from the ILEP. The ML application may receive input data from the ILU, from the host device, or from another network entity (e.g., from a network device such as a sensor, a data producer, etc.). The ML application may generate output data that may be used by the host device or another network entity. The host device may be located within a communication network so that the ML-based application may access the required input data and may output data for the relevant network entity.

[0095] Exemplary network entities include, but are not limited to:

[0096] - Radio access network entities, such as base stations (e.g., cellular base stations such as eNodeBs in LTE and LTE-Advanced networks and gNodeBs used in 5G networks, and femtocells used in homes or business centers);

[0097] -Relay station;

[0098] - Control stations (e.g., radio network controllers, base station controllers, network switching subsystems);

[0099] - Access points in local area networks or ad hoc networks;

[0100] - Gateway and radio access network entities;

[0101] - Network management entities (e.g., operations, maintenance and administration (OAM) entities);

[0102] -Network automation system;

[0103] - Distributed analysis entities, such as autonomous systems (D-SON);

[0104] - Network functions (e.g., Network Data Analysis Function or NWDAF as defined in current 3GPP standards).

[0105] One or more ILUs may be stored in an intent logic library (ILL). The ILUs are stored in association with ILU capability information derived from capability information received for one or more corresponding ML-based applications mapped to the ILU. The ILL may be configured to perform an ILU search based on search criteria associated with the capability information.

[0106] The ILU may be stored in the ILL in association with at least one of an identifier, a name, and a description that can be understood by a user to allow identification of the ILU. For example, each ILU has a description for a human user who may want to modify, reuse, or delete the ILU.

[0107] In one or more embodiments, the proposed system provides an ML capability discovery process implemented between two functions: an ML capability consumer and an ML capability producer.

[0108] ML Capability Consumer is a function that can be implemented by the Intent Logic Execution Platform or Intent Realization System or any other entity configured to generate new ILU or any other execution logic concept. ML Capability Producer is a function that can be implemented by the ML-based application itself or by the ML coordinator or a dedicated network entity (e.g., network management function, automation function, analytics function or network function such as gNB, or cell) that is configured to provide ML capability exposure interface on behalf of one or more ML-based applications or one or more ML coordinators.

[0109] Authorization may be required to allow one or more ML capability consumers to request capabilities, so that only authorized ML capability consumers can request capabilities of existing ML-based applications. Likewise, ML-based applications can only report their capabilities to authorized ML capability consumers.

[0110] Authorized ML capability consumers (typically intent implementation systems or ILEPs) can be configured to request mapping of specified ILUs to a set of applicable ML-based applications from the ILL (Intent Logic Library). ILL can be configured to report mapping of specified ILUs to a set of applicable ML-based applications.

[0111] Capability information may be disclosed through an interface (e.g., an open interface) between an ML capability consumer and an ML capability producer. The interface may be used for various operations, including at least one of the following:

[0112] - Sending a request for capability information from an ML capability consumer to an ML capability producer;

[0113] - receiving capability information from ML capability producers by ML capability consumers;

[0114] -Send configuration data of ML applications from ML capability consumers to ML capability producers.

[0115] The ML capability consumer receives capability information representing capabilities of one or more machine learning (ML) based applications. The capability information may be received in response to a request sent from the ML capability consumer to one or more ML capability producers (e.g., to each new ML based application that is available). The capability information may also be received according to a subscription scheme, wherein the ML capability consumer sends at least one request (e.g., to a network entity that implements ML capability producer functionality on behalf of ML based applications) to receive capability information, and receives capability information matching its request when new ML applications are available, without having to send a request to each new ML based application.

[0116] There may be several deployment / configuration options for mapping the ILU to an ML-based application and using the mapped ILU to implement intents, such as Figure 3 and Figure 4 as shown in .

[0117] exist Figure 3 In the example of , the ML capability producer is the ML application 261 and the ML capability consumer is the ILEP 230. The capability information may be exposed through an interface 265 between the ML capability consumer and the ML capability producer.

[0118] exist Figure 3 In the example of ILL 235, an ILU 251 in ILL 235 is mapped to an ML-based application 261. The ML-based application 261 may implement one or more ML models. ILEP 230 may load ILU 251 from a database and execute ILU 251, which itself initiates the execution of ML application 261.

[0119] exist Figure 4 In the example of FIG. 2 , several ML capability producers are provided, including: ML applications 271 to 273, ML applications 274 to 276, and ML coordinator 280. The ML capability consumer is ILEP 230. Figure 3 Similarly, each ML-based application can implement one or more ML models.

[0120] The capability information is exposed through interfaces 265-1 to 265-4 between the ML capability consumer (ILEP 230) and the ML capability producer (here, ML applications 271 to 273 and ML coordinator 280). The interfaces 265-1 to 265-3 are implemented between the ILEP 230 and each ML application 271 to 273. The interface 265-4 is implemented between the ILEP 230 and the ML coordinator 280, without implementing an interface between the ILEP 230 and the ML applications 274 to 276 coordinated by the ML coordinator 280.

[0121] exist Figure 4 In the example of , one ILU is mapped to several ML-based applications. There are several options to map the ILU to several ML-based applications.

[0122] In the first option, the ML capability consumer (ILEP 230) derives an ILU 252 from the ML-based applications 271 to 273 without overlooking the ML-based applications (i.e., without the ML coordinator), so that the ILU 252 is mapped with the ML-based applications 271 to 273. The ILEP 230 may load the ILU 252 from the ILL 235 and execute the ILU 252.

[0123] In this first option, ILU 252 may initiate the execution of one or more ML applications mapped with the ILU. For example, ILU 252 initiates the execution of each of ML applications 271 to 273. Alternatively, ILU 252 may be configured to initiate one of the ML applications mapped with this ILU (e.g., ML application 271), and one of the ML applications itself may be configured to initiate at least one other ML application mapped with this ILU (e.g., ML applications 272 and / or 273). In both alternatives, ML applications 271, 272, 273 may be executed in a sequential manner, or at least partially in parallel or in parallel.

[0124] In the second option, the ML capability consumer (ILEP 230) derives an ILU 253 from the ML-based applications 274 to 276 and an overhead ML application (i.e., ML coordinator 280) responsible for coordinating the execution of the ML-based applications, such that the ILU 253 is mapped with the ML-based applications 274 to 276 and the ML coordinator 280.

[0125] In this second option, the ILU 253 initiates the execution of one or more ML applications mapped to the ILU, the execution and performance of these ML applications being monitored by the ML coordinator 280. For example, the ILU 252 initiates the execution of the ML coordinator 280, which in turn initiates the execution of the ML-based applications 274 to 276 coordinated by the ML coordinator 280. In this second option, the ML applications 274, 275, 276 may be executed in a sequential manner, or at least partially in parallel or in parallel.

[0126] The capability information disclosed for available ML functions may include various types of information. The capability information derived for the ILU may include the same types of information.

[0127] The capability information of the ML-based application may include a description of the functions or ML models implemented by the relevant ML-based application. The description may include at least one of text (e.g., "predicting QoS level within a QoS range"), keywords, identification of functions within a predefined function list, and the like.

[0128] The capability information of the ML-based application may include information indicating the type of function performed by the ML-based application for the communication network. The type may be at least one of: an optimization function, a control function, an analysis function, and an ML coordination function. Other relevant types may be used. The function type may be identified by an identifier or name in an identifier list, or identified using another type of identification method.

[0129] The capability information of the ML-based application may include information representing entities to which the functions implemented by the ML-based application are applied. These entities may be one or more objects or one or more object types (e.g., "UE" indicates that an object is a user equipment). An object may be used as an input or output of an ML-based application. An object may correspond to any entity in a communication network: a UE, a network entity, a functional unit, a virtual function, a database, or other similar / related object. An object or object type may be identified by a name or an identifier, or using another type of identification method.

[0130] The capability information may include information indicating at least one configuration parameter of an object or object type that performs the function. The configuration parameter of an object or object type may be any parameter that can be used to configure the relevant object. For example, a cell needs to have an antenna tilt to determine where the antenna should be pointed in order to maximize coverage and minimize interference. The antenna tilt in this example is a configuration parameter of the object cell. The configuration parameter of an object or object type is an example of a configuration parameter of an ML-based application.

[0131] The capability information may include information representing configuration parameters for an ML model implemented by an ML-based application. Configuration parameters (meta-parameters, input parameters, etc.) of an ML model may be any parameters that can be used to configure the associated ML model. For example, an ML model may have inputs that the ML model may use to drive decisions, e.g., an ML model for optimizing coverage may use antenna tilt as input data. In this case, antenna tilt is a configuration parameter of the ML model. Configuration parameters for objects or object types are examples of configuration parameters of an ML-based application.

[0132] The capability information may include information representing goals and / or constraints for the performance of the function, wherein the goals and / or constraints are defined based on at least one network metric. Such goals and / or constraints may be encoded in text, may be identified by an identifier or name within a goal list, or may be identified using another type of identification method.

[0133] The capability information of the ML-based application can be encoded and reported in various formats. The capability information of the ML-based application can be encoded using descriptive text and / or triples and / or tables, etc.

[0134] Capability information can represent a decision that can be reported. For example, capability information can be a triple<x,y,z> in the form of

[0135] -x: objects or object types that can be optimized or controlled by ML-based applications;

[0136] -y: configuration parameters of object or object type x, based on ML application optimization control to achieve the desired results;

[0137] -z: The network metric that the ML-based application optimizes through its actions.

[0138] Capability information can represent the analysis that can be reported. For example, in the form of a tuple<x,z> Instructions in the form of:

[0139] -x: Objects or object types that can be analyzed by ML-based applications

[0140] -z: Network characteristics (about object x) for which analysis is generated based on the application of ML.

[0141] Figure 5 is a flow chart illustrating an exemplary implementation of a method intended to be implemented in accordance with some embodiments.

[0142] The method is implemented by an ML capability consumer 550 (eg, an intent logic execution platform) and an ML capability producer 590 and an intent specification platform 520 .

[0143] ML capability consumers 590 discover available ML-based applications and their supported capabilities provided by ML capability producers. ML capability consumers may send (step 501) a request for ML capability information and receive a response (step 502) including ML capability information from ML capability producers 590. Rather than sending ML capability information in response to a request, ML capability consumers 530 receive ML capability information from one or more ML capability producers 590 when one or more new ML-based applications are available, e.g., based on a subscription plan and / or without sending a request. This discovery phase (steps 501 and 502) may be repeated as needed.

[0144] The ML capability consumer 530 derives (step 503) one or more new ILUs from one or more ML-based applications based on the received ML capability information. The ML capability consumer stores each new ILU in a library (e.g., intent logic library, ILL) along with the ILU capability information and mapping information (step 503). The mapping information may include an identification of the ML-based application mapped to the ILU. The ILU capability information of the ILU may be derived from the ML capabilities of the ML-based application mapped to the ILU.

[0145] The ILL may be configured to provide capability information for one or more ILUs and / or mapping information of a mapping between an ILU and one or more ML-based applications. The ILL may be configured to perform an ILU search based on search criteria related to the capability information and / or mapping information of the ILU to be found.

[0146] The ML capability consumer receives (step 504) an intent (e.g., from an operator) and identifies (step 505) in the intent logic library the appropriate ILU(s) mapped with the ML-based application to implement the received user intent. The ILU(s) may be identified based on ILU capability information stored with each ILU. It should be noted that non-ML-based logic and / or commands (in addition to the ML-based ILUs) may still be required to implement a given intent.

[0147] Patent application published under number WO2021 / 213632A1 (incorporated herein by reference) discloses an embodiment of a system for implementing user intent using intent logic units (ILUs) stored in an intent logic library (ILL). For a given intent to be implemented, one or more intent logic units may be used and executed to achieve the intent.

[0148] The ILEP 230 may search the ILL 235 for an ILU 250 that matches the specified intent. The search may be based on the capability information and / or description and / or mapping information of the ILU. For a given ILU, the description of the ILU may include at least one of the following: descriptive text, one or more functions of the ILU, one or more parameters of the ILU, etc.

[0149] If the ILU exists, ILL 235 returns the ILU 250 to ILEP 230, otherwise a failure event is returned. If the ILU 230 exists, ILEP 230 can schedule the execution of ILU 250, i.e., the ILEP can check whether the ILU can be executed immediately, or whether it must wait and must be scheduled at a different time. After executing the identified ILU or sequence of ILUs, the ILEP can evaluate the results of the ILU execution to confirm that the intent has been achieved.

[0150] If the ILU does not exist, ILEP 230 may determine which ILUs may be combined to achieve the specified intent. If a combined ILU may be generated by combining several ILUs, ILEP 230 may schedule the execution of the combined ILU. Otherwise, ILEP 230 may return an execution failure to ISP 220.

[0151] Once the appropriate ILU(s) are identified in step 205, the ML capability consumer may configure and / or execute (step 506) the identified ILU(s), thereby initiating execution of an ML-based application mapped to the identified ILU(s) to implement the received intent.

[0152] Figure 6 An example flow chart of a method for intent implementation using an ML-based solution is shown. The method involves multiple entities: an intent specification platform 620, an intent logic library 635, an intent logic execution platform 630, and an ML capability producer 690, which includes at least one ML orchestrator 680 and at least one ML application 660.

[0153] Step 601 : Discovery phase: The intention logic execution platform discovers available ML-based applications, such as the ML coordinator 680 and the ML application 660 .

[0154] Steps 602 and 603: Request capability information. The intent logic execution platform 630 requests detailed information about the capabilities it supports from one or more ML capability producers 690 (ML-based applications 660 and / or ML orchestrators 680). Each request for capability information may be sent to an ML-based application 660 (step 603) or an ML orchestrator 680 (step 602).

[0155] Steps 604 and 605: Publishing of capabilities. One or more ML capability producers 690 (ML-based applications 660 and / or ML coordinators 680) may report their capabilities. In step 604, the ML coordinator 680 may send capability information related to the ML applications coordinated by the ML coordinator 680. In step 605, the ML application 660 may send its own capability information.

[0156] The capability information received for the ML application 660 may be associated with an identifier of an ML model implemented by the ML-based application 660 (eg, ModelID1) or an identifier of the ML-based application itself. The capability information received for the ML orchestrator 680 may be associated with an identifier of the ML orchestrator (eg, MLOrchestratorID1).

[0157] For example, the capability information of ML-based applications includes: ModelID1, QoS level within the predicted QoS range, [Ues, QoS ID (QCI / 5QI), QoS target]

[0158] in

[0159] - "ModelID1" is an identifier of an ML-based application or an associated ML model,

[0160] - "QoS level within the predicted QoS range" is capability information related to the function performed by the ML application, and

[0161] - “[Ues, QoS ID (QCI / 5QI), QoS target]” defines the object type (here UE and QoS parameters) to which the functions performed by the ML model are applied.

[0162] Similarly, as another example, the capability information of the ML-based application includes:

[0163] ModelID2, "Reduce the switching indicator by percentage P"

[0164] in

[0165] - "ModelID2" is an identifier of an ML-based application or an associated ML model,

[0166] - "Reduce the handover index by percentage P" is capability information related to the function performed by the ML application, and

[0167] The handover indicators include: ping-pong, radio link failure due to too early handover, and radio link failure due to too late handover.

[0168] Similarly, as another example, the capability information of the ML-based application includes:

[0169] ModelID3, "Move users from cell / frequency layer X to cell / frequency layer Y by reducing the cell or layer load below threshold T".

[0170] Where X, Y, and T are configuration parameters of ML-based applications

[0171] Similarly, as another example, the capability information of the ML-based application includes:

[0172] ModelID4, "Predict the user's position, speed, and trajectory at time instance X or time period Y to Z"

[0173] Where X, Y, and Z are configuration parameters for ML-based applications

[0174] Similarly, as another example, the capability information of the ML coordinator includes:

[0175] MLOrchestratorID1, "Ensures the coordination of models with the following IDs: ModelID3, ModelID4, supports performance monitoring of MLQoS metrics: KPI1, KPI2"

[0176] Step 606: Create an ILU mapped to an ML-based application. The intent logic execution platform derives one or more ILUs based on the discovered ML-based application capability information received in steps 604 and 605. In addition, the intent logic execution platform may store an ILU identifier associated with the ILU capability information derived from the capability information of the one or more ML applications mapped to the ILU. In addition, the intent logic execution platform may store mapping information for mapping between the ILU and the corresponding one or more ML-based applications mapped to the ILU.

[0177] For example, for an ILU mapped with an ML-based application identifier via ModelID1, the following information can be stored

[0178] ILU1, "Ensure that target QoS is met", MAPPING(ModelID1, ILU1)

[0179] in

[0180] - "ILU1" is the identifier of the ILU;

[0181] - "Ensure that the target QoS is met" is the capability information of the ILU;

[0182] - "MAPPING(ModelID1, ILU1)" is mapping information defining the mapping between the ILU identified by ILU1 and the ML model identified by ModelID1.

[0183] Similarly, as another example, for an ILU mapped with an ML-based application identifier via ModelID2, the following information may be stored:

[0184] ILU2, "Minimizing RLF due to incorrect switching decisions", MAPPING (ModelID2, ILU2)

[0185] Similarly, as another example, for an ILU mapped with an ML-based application identifier via ModelID3, the following information may be stored:

[0186] ILU3, "Move user from technology / cell / frequency layer X to technology / cell / frequency layer Y", MAPPING(ModelID3, ILU3)

[0187] Similarly, as another example, for an ILU mapped with an ML-based application identifier via ModelID4, the following information may be stored:

[0188] ILU4, "Clear cell layer X", MAPPING (ModelID3, ILU4)

[0189] Similarly, as another example, for an ILU mapped with an ML-based application identifier via ModelID5, the following information may be stored:

[0190] ILU5, "Inferring the location of a user at a specific time instance or period", MAPPING (ModelID5, ILU5)

[0191] Step 607: The intent logic execution platform may create an ILU mapped to several ML applications in order to combine their capabilities. There are several ways to combine the capabilities of ML applications.

[0192] In a first embodiment, the intent logic execution platform generates a new ILU mapped to several ML applications. For example, the ILU is mapped to a specific combination of these ML applications, according to which the ML applications are executed in a sequential manner (without using an ML coordinator), so that the output data of one ML application is used as the input data of a subsequent ML application.

[0193] In a second embodiment, the intent logic execution platform exports the combined ILU as an ILU combination, so that the combined ILU combines the capabilities of the ML application mapped to the combined ILU. This more complex ILU will also be mapped to several ML-based applications required to implement the complex ILU.

[0194] In a third embodiment, the intent logic execution platform exports an ILU that is mapped to the ML orchestrator, such that the newly exported ILU combines the capabilities of the ML application orchestrated by the ML orchestrator.

[0195] For example, for an ILU mapped to an ML application identified by ModelID1 and ModelID3, the following information may be stored in association with the ILU:

[0196] ILU6, "Use technology / cell / frequency layer X for users with specific QoS conditions", MAPPING (ModelID1, ModelID3, ILU6)

[0197] in

[0198] - "ILU6" is the identifier of the ILU;

[0199] - "Technology / cell / frequency layer X for users that meet certain QoS conditions" is capability information that defines the functionality of the ILU;

[0200] - "MAPPING(ModelID1, ModelID3, ILU6)" is mapping information defining the mapping between the ILU identified by ILU6 and the ML applications identified by ModelID1 and ModelID3.

[0201] Likewise, for an ILU mapped to an ML application identified by ModelID3 and ModelID4, the following information may be stored in association with the ILU:

[0202] ILU7, "Clear technology / cell / frequency layer X in a specific area at a specific time for a user", MAPPING(ModelID3, ModelID4, ILU7)

[0203] Likewise, for an ILU identified by ILU8 mapped with an ML application coordinated by an orchestrator identified by identifier MLOrchestratorID1, the following information may be stored in association with the ILU:

[0204] ILU8, “Monitor and ensure ML model accuracy is above X%”, MAPPING (MLOrchestratorID1, ILU8)

[0205] Step 608: For each newly derived ILU, the intent logic execution platform stores the capability information of the newly derived ILU and its mapping to the corresponding ML-based application in association with the corresponding identifier of the newly derived ILU.

[0206] The capability information of the newly derived ILU is derived from the capability information of the corresponding (multiple) ML-based applications mapped to the ILU. When the new ILU is mapped to a single ML-based application, the capability information of the newly derived ILU may be the same as the capability information of the ML-based application. When the new ILU is mapped to several ML-based applications, the capability information of the newly derived ILU corresponds to the capability information of all mapped ML-based applications. The capability information of the newly derived ILU may be automatically generated and / or checked and / or approved by a user (operator).

[0207] Step 609: The intent logic execution platform receives a request to implement a new intent. Users can request to implement different types of intents. For example:

[0208] - Intent A: “QoS target deteriorates by X% of users moving to the 5G layer”;

[0209] - Intent B: “Clear the 5G layer in rural areas at night so that the 5G layer can be turned off and then the corresponding cells can be turned off. Ensure the accuracy of the solution is at least 90%”.

[0210] Step 610: The intent logic execution platform searches for and selects an ILU in the ILL that matches the received intent. The search in the ILL may include semantic analysis and / or keyword analysis of the capability information of the ILUs stored in the ILL based on the specified intent. The selection may be checked and / or approved by the user (operator). If several matching ILUs are found in the ILL, the selection of the ILU to be used may be finally executed and / or approved by the user (operator).

[0211] For example:

[0212] - To implement intent A, select the ILU identified by ILU6 that maps to (ModelID1, ModelID3);

[0213] - To implement intent B, select the ILU identified by ILU7 that maps to (ModelID3, ModelID4) and the ILU identified by ILU8 that maps to (MLOrchestratorID1);

[0214] Step 611: The intent logic execution platform initiates execution of ML-based applications using the one or more ILUs selected in step 610, wherein each selected ILU initiates one or more ML-based applications mapped to the associated ILU. Configuration parameters extracted from the intent to be implemented may be used by the ILEP or by the ILU selected in step 610 to configure one or more ML-based applications.

[0215] For example, to achieve intention A above:

[0216] - The configuration parameters extracted from Intent A include the level of QoS target reduction (X%) and the indication of the frequency layer (e.g. from any frequency layer to 5G frequency layer); these configuration parameters are used by the ILEP to configure the ML applications ModelID1 and ModelID3;

[0217] -ILEP performs the mapping to the selected ILU identified by ILU6 (ModelID1, ModelID3);

[0218] -ILU ILU6 triggers the execution of ML-based applications that implement ML applications ModelID1 and ModelID3;

[0219] -ML application ModelID1 is configured to detect users whose QoS target experience drops by X%;

[0220] -ML application ModelID3 is configured to move users detected by ML application ModelID1 from any frequency layer to the 5G frequency layer.

[0221] Steps 612, 613, and 614: The intent logic execution platform uses the one or more ILUs selected in step 610 to initiate execution of the ML-based application 660 (step 613) and the ML coordinator 680 (step 612), wherein each selected ILU initiates one or more ML-based applications mapped to the associated ILU. The ML coordinator 680 monitors (step 614) the performance of the initiated ML-based application 660.

[0222] For example, to implement the above intent B, ILEP extracts configuration parameters from intent B, including: the location of the relevant user (e.g., urban area coordinates), the required time (e.g., "any time instance"), an indication of the frequency layer (e.g., from the 5G frequency layer to any frequency layer), and accuracy (e.g., 90%). These configuration parameters will be used by ILEP to configure ML applications ModelID4, ModelID3, and ML coordinator MLOrchestratorID1. Then:

[0223] -ILEP executes the selected ILUs, namely ILU7 mapped to the ML application (ModelID3, ModelID4) and ILU8 mapped to the ML orchestrator (MLOrchestratorID1) for monitoring the ML application (ModelID3, ModelID4)

[0224] - ILU8 triggers the execution of ML coordinator MLOrchestratorID1 (step 612);

[0225] - ILU7 triggers the execution of the ML application (ModelID3, ModelID4) (step 613);

[0226] Once the ML coordinator MLOrchestratorID1 is triggered by ILU8, it will monitor the execution of ModelID3 and ModelID4 (step 614);

[0227] - ILEP executes configuration of the ML application ModelID4 so that it identifies users within the urban area at any time instance (e.g., based on urban area coordinates);

[0228] -ILEP executes the configuration of ML application ModelId3 so that it moves the user identified by ModelID4 from the 5G frequency layer to any frequency layer;

[0229] -ILEP executes the configuration of ML OrchestratorID1 to monitor and ensure that the accuracy of the monitored ML models (i.e. ModelID3 and ModelID4) is not less than 90%.

[0230] Applying ML-based solutions to intent B may enable the conditions required to shut down the cell (clear the frequency layer) to be achieved. However, in addition to ML-based solutions, the full intent implementation may also require non-ML-based commands, such as actual cell shut-down, which may be performed automatically or with operator assistance.

[0231] Figure 7 A flow chart of a method for intent implementation is shown. The method involves the ML capability consumer disclosed herein and the ML capability producer disclosed herein.

[0232] In step 710: the ML capability consumer may send a request for capability information to at least one ML capability producer. The request may be sent to at least one of: an ML-based application, an ML application coordinator, and another ML capability producer. The request may be a request to receive responsive capability information for currently available capabilities, or a subscription request to receive (e.g., on a periodic basis) capability information for currently available capabilities and capabilities that may be available in the future.

[0233] In step 720 , an ML capability consumer receives capability information representing capabilities of one or more machine learning (ML) based applications (eg, from one or more at least one ML capability producer).

[0234] In step 730, the ML capability consumer generates one or more intent logical units (ILUs). The newly created ILU is mapped with at least one corresponding ML-based application and is configured to initiate execution of the at least one corresponding ML-based application mapped with the ILU.

[0235] In step 740, the ML capability consumer stores one or more ILUs in the intent logic library. Each ILU is stored in association with capability information derived for the ILU from capability information received for at least one corresponding ML-based application mapped to the ILU. The ML capability consumer may store mapping information for mapping performed between the ILU and at least one corresponding ML-based application in the intent logic library in association with the stored ILU.

[0236] In step 750, the ML capability consumer receives a request including intent information representing an intent to be implemented with respect to at least one communication network.

[0237] In step 760, the ML capability consumer identifies one or more ILUs configured to initiate execution of one or more corresponding ML-based applications suitable for implementing the intent for the at least one communication network based on the ILU capability information stored in the intent logic library.

[0238] In step 770, the ML capability consumer executes the identified one or more ILUs. Execution of the identified one or more ILUs may result in initiating execution of one or more ML-based applications mapped to the identified one or more ILUs. Initiating may include configuring the mapped ML-based application(s) with one or more configuration parameters. The one or more configuration parameters for the ML-based application may be received from the ILU mapped to the ML-based application or from the ML capability consumer (e.g., from an ILEP).

[0239] Figure 8A flow chart of a method for implementing an intent is shown. The method involves the ML capability consumer disclosed herein and the ML capability producer disclosed herein.

[0240] In step 810, an ML capability producer receives a request for capability information (eg, from an ML capability consumer such as an ILEP).

[0241] In step 820, the ML capability producer sends capability information representing the capabilities of one or more ML-based applications in response to the request. If the ML capability producer is an ML coordinator, the ML capability producer sends capability information representing the capabilities of one or more ML-based applications monitored by the ML coordinator. If the ML capability producer is an ML-based application, the ML capability producer sends capability information representing the capabilities of the ML-based application. If the ML capability producer is a network entity that implements the ML capability producer function on behalf of one or more ML-based applications, the ML capability producer may send capability information representing the capabilities of these one or more or all ML-based applications.

[0242] In step 830, the host device executes one or more ML-based applications. The execution is initiated by one or more ILUs mapped to the one or more ML-based applications. The one or more ILUs are selected and executed to achieve a given intent. The initiation may include configuring the one or more mapped ML-based applications using one or more configuration parameters by each relevant ILU. The configuration parameters may be stored in the ILU so that when a given ILU is selected, the configuration parameters stored by the ILU are selected).

[0243] The initiation may be performed by the ML coordinator. In this case, the ML coordinator may receive (e.g., from an ML consumer or from an ILU mapped with the ML coordinator) one or more configuration parameters for one or more ML-based applications monitored by the ML coordinator. The ML coordinator itself may configure the one or more ML-based applications monitored by the ML coordinator with the received one or more configuration parameters before and / or when and / or after triggering execution of the one or more ML-based applications monitored.

[0244] The proposed ML-based intent implementation solution is related to an intent-driven management service for a mobile network and provides a method for converting received intents into executable actions. The intent-driven management service may correspond to a management service that allows a user (e.g., a consumer of the intent-driven management service) to express an intent. The intent-driven management service allows the user to express a desired intent for managing a network and services. The intent-driven management service producer interprets the expressed intent to convert the intent into an executable task.

[0245] The proposed solution is also related to the management of the ML functionality, in particular the configuration of the ML functionality. The proposed solution is also related to creating a new relationship between the management function of the ML functionality and other entities within the communication network, such as the gNB.

[0246] ML applications may be configured to implement a given analytical and / or decision-making functionality that may be used in the context of a variety of applications and use cases in different systems and various industries.

[0247] Those skilled in the art should understand that any functions, engines, block diagrams, flow charts, state transition diagrams, flowcharts and / or data structures described herein represent conceptual views of illustrative circuit systems that embody the principles of the present invention. Similarly, it should be understood that any flow charts, flow charts, state transition diagrams, pseudocodes, etc. represent various processes that can be substantially represented in a computer-readable medium and executed by a computer or processing device, regardless of whether such a computer or processor is explicitly shown.

[0248] Although a flow chart may describe operations as a sequential process, many operations may be performed in parallel, concurrently, or simultaneously. In addition, some operations may be omitted, combined, or performed in a different order. A process may be terminated when its operations are completed, but may also have additional steps not disclosed in the figure or description. A process may correspond to a method, function, procedure, subroutine, subprogram, etc. When a process corresponds to a function, its termination may correspond to the function returning to the calling function or the main function.

[0249] Each function, engine, block, step described herein can be implemented in hardware, software, firmware, middleware, microcode, or any suitable combination thereof.

[0250] When implemented in software, firmware, middleware, or microcode, instructions for performing the necessary tasks may be stored in a computer-readable medium, which may or may not be included in a host device or system. Instructions may be transmitted via a computer-readable medium and loaded onto a host device or system. Instructions are configured to cause the host device / system to perform one or more functions disclosed herein. For example, as mentioned above, according to one or more examples, at least one memory may include or store instructions, and at least one memory and instructions may be configured together with at least one processor to cause the host device / system to perform one or more functions. In addition, processors, memories, and instructions are used as components that provide or cause the host device / system to perform one or more functions disclosed herein.

[0251] The host device or system may be a general purpose computer and / or computing system, a special purpose computer and / or computing system, a programmable processing device and / or system, a machine, etc. The host device or system may be or include or be part of a user device, a client device, a mobile phone, a laptop, a computer, a network element, a data server, a network resource controller, a network device, a router, a gateway, a network node, a computer, a cloud-based server, a web server, an application server, a proxy server, etc.

[0252] Fig. 9 An example embodiment of an apparatus 1000 is illustrated. The apparatus 1000 may be a host apparatus or part of a host system disclosed herein. The apparatus may be configured to host at least one entity disclosed herein: an ML capability consumer function, an ML capability producer function, one or more ML-based applications, an ILEP or one or more functions of an ILEP, an ILL or one or more functions of an ILL.

[0253] like Fig. 9 Schematically represented, the device 1000 may include at least one processor 1010 and at least one memory 1010. The device 1000 may include one or more communication interfaces 1040 (e.g., a network interface for accessing a wired / wireless network, including an Ethernet interface, a WIFI interface, a USB interface, etc.), which are connected to the processor and configured to communicate via (multiple) wired / wireless communication links. The device 1000 may include other related hardware connected to the processor, such as a user interface 1030 (e.g., a keyboard, a mouse, a display screen, etc.). The device 1000 may also include one or more media drives 1050 for reading computer-readable storage media (e.g., a digital storage disk 1060 (CD-ROM, DVD, Blu-ray, etc.), a USB key 1080, etc.). The processor 1010 is connected to each of the other components 1030, 1040, 1050 to control their operation.

[0254] The memory 1020 may include random access memory (RAM), cache, non-volatile memory, backup memory (e.g., programmable or flash memory), read-only memory (ROM), hard disk drive (HDD), solid state drive (SSD), or any combination thereof. The ROM of the memory 1020 may be configured to store one or more computer program codes of an operating system and / or one or more software applications of the storage device 1000, etc. The RAM of the memory 1020 may be used by the processor 1010 to temporarily store data.

[0255] The processor 1010 can be configured to store, read, load, execute and / or otherwise process instructions 1070 stored in computer-readable storage media 1060, 1080 and / or memory 1020, so that when the instructions are executed by the processor, the device 1000 performs one or more or all steps of the method described herein for the relevant device 1000.

[0256] Instructions can correspond to computer program instructions, computer program codes and can include one or more code segments. Code segments can represent any combination of processes, functions, subprograms, programs, routines, subroutines, modules, software packages, classes or instructions, data structures or program statements. Code segments can be coupled to another code segment or hardware circuit by transmitting and / or receiving information, data, arguments, parameters or memory contents. Information, arguments, parameters, data, etc. can be transmitted, forwarded or transmitted via any suitable technology, including memory sharing, message passing, token passing, network transmission, etc.

[0257] When provided by a processor, the functionality may be provided by a single dedicated processor, by a single shared processor, or by multiple individual processors, some of which may be shared. The term "processor" should not be interpreted as referring only to hardware capable of executing software, and may implicitly include one or more processing circuits, whether programmable or not. A processor or similar processing circuit may correspond to a digital signal processor (DSP), a network processor, an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), a system on a chip (SoC), a central processing unit (CPU), a processing unit (CPU), an arithmetic logic unit (ALU), a programmable logic unit (PLU), a processing core, programmable logic, a microprocessor, a controller, a microcontroller, a microcomputer, any device capable of responding and / or executing instructions in a defined manner and / or according to a defined logic. Other conventional or customized hardware may also be included. A processor or processing circuit may be configured to execute instructions suitable for causing a host device or system to perform one or more functions disclosed herein for a related host device or system.

[0258] Computer readable media or computer readable storage media can be any tangible storage medium suitable for storing instructions readable by a computer or processor. Computer readable media can more generally be any storage medium capable of storing and / or containing and / or carrying instructions and / or data. Computer readable media can be portable or fixed storage media. Computer readable media can include one or more storage devices, such as permanent mass storage devices, magnetic storage media, optical storage media, digital storage disks (CD-ROM, DVD, Blu-ray, etc.), USB keys or dongles or peripheral devices, memory suitable for storing instructions readable by a computer or processor.

[0259] A memory suitable for storing computer or processor readable instructions may be, for example, a read-only memory (ROM), a permanent mass storage device such as a disk drive, a hard disk drive (HDD), a solid-state drive (SSD), a memory card, a core memory, a flash memory, or any combination thereof.

[0260] In this specification, the wording "component configured to perform one or more functions" or "component for performing one or more functions" may correspond to one or more functional blocks, which include a circuit system suitable for performing or configured to perform (multiple) related functions. The block may perform this function by itself, or may cooperate and / or communicate with other one or more blocks to perform this function. "Component" may correspond to or be implemented as "one or more modules", "one or more devices", "one or more units", etc. The component may include at least one processor and at least one memory including computer program code, wherein the at least one memory and the computer program code are configured to cause the device or system to perform (multiple) related functions together with the at least one processor.

[0261] As used in this application, the term "circuitry" may refer to one, more, or all of the following:

[0262] (a) pure hardware circuit implementation (such as implementation in analog and / or digital circuitry only) and

[0263] (b) a combination of hardware circuitry and software, such as (as applicable): (i) a combination of analog and / or digital hardware circuitry and software / firmware; and (ii) any portion of hardware processor(s) with software (including digital signal processor(s), software and memory that work together to enable a device such as a mobile phone or server to perform various functions); and

[0264] (c) hardware circuit(s) and / or processor(s), such as microprocessor(s) or portion(s) of microprocessor(s), that require software (e.g., firmware) for operation, but where the software is not required for operation, the software may not be present.”

[0265] This definition of circuitry applies to all uses of the term in this application, including in any claims. As a further example, as used in this application, the term circuitry also covers an implementation of only a hardware circuit or processor (or multiple processors) or a portion of a hardware circuit or processor and its (or their) accompanying software and / or firmware. By way of example and where applicable to a particular claim element, the term circuitry also covers an integrated circuit for a network element or network node or any other computing device or network device.

[0266] The term circuit system may encompass digital signal processor (DSP) hardware, network processors, application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), etc. The circuit system may be or include, for example, hardware, programmable logic, a programmable processor that executes software or firmware, and / or any combination thereof (e.g., a processor, a control unit / entity, a controller) to execute instructions or software and control the transmission and reception of signals, and a memory for storing data and / or instructions.

[0267] The circuit system may also make decisions or determinations, generate frames, packets, or messages for transmission, decode received frames or messages for further processing, and other tasks or functions described herein. The circuit system may control the transmission of signals or messages over a radio network, and may control the reception of signals or messages via one or more communication networks, etc.

[0268] Although the terms first, second, etc. can be used to describe various elements in this article, these elements should not be limited by these terms. These terms are only used to distinguish one element from another element. For example, the first element can be called the second element, and similarly, the second element can be called the first element without departing from the scope of this disclosure. As used herein, the term "and / or" includes any and all combinations of one or more associated listed items.

[0269] The terms used herein are only used to describe specific embodiments and are not intended to be limiting. As used herein, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are also intended to include the plural forms. It should be further understood that the terms "comprise" and / or "include" when used herein specify the presence of stated features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof.

[0270] Although various aspects of the present disclosure have been specifically shown and described with reference to the above embodiments, it will be appreciated by those skilled in the art that various additional embodiments may be conceived by modifying the disclosed machines, systems, and methods without departing from the disclosed scope. It should be understood that these embodiments fall within the scope of the present disclosure as determined based on the claims and any equivalents thereof.

[0271] List of abbreviations

[0272] 5G Fifth Generation

[0273] CAN Cognitive Autonomous Network

[0274] CF cognitive function

[0275] CLI Command Line Interface

[0276] CNM Cognitive Network Management

[0277] GUI Graphical User Interface

[0278] IBN Intent-Based Networking

[0279] IoT

[0280] KPI Key Performance Indicator

[0281] NE Network Element

[0282] NM Network Management

[0283] OAM operations, maintenance and management

[0284] SON Self-Organizing Network

Claims

1. An apparatus comprising means for performing a method, the method comprising: sending a request for capability information, wherein the request is sent to at least one of: an ML-based application, an ML application coordinator, and an ML capability producer; Receiving capability information indicating capabilities of one or more machine learning (ML) based applications; generating one or more intent logic units (ILUs), wherein the ILUs are mapped with at least one corresponding ML-based application, wherein the ILUs are configured to initiate execution of the at least one corresponding ML-based application mapped with the ILUs; The one or more ILUs are stored in an intent logic library, wherein the ILUs are stored in association with capability information derived for the ILU from capability information received for the at least one corresponding ML-based application mapped with the ILU.

2. The apparatus of claim 1, wherein the capability information received for an ML-based application comprises: Information representing a type of function performed by the ML-based application for at least one communication network, wherein the type is at least one of: an optimization function, a control function, an analysis function, and an ML coordination function.

3. The apparatus according to claim 1 or 2, wherein the ILU is configured to initiate execution of the at least one corresponding ML-based application and / or initiate an ML coordination function configured to monitor the at least one corresponding ML-based application.

4. The apparatus of claim 2 or 3, wherein the capability information received for the ML-based application comprises: Information indicating one or more objects or one or more object types that perform the described function.

5. The apparatus of any one of claims 2 to 4, wherein the capability information received for an ML-based application comprises: Information representing at least one configuration parameter of an object or object type that performs the function.

6. The apparatus of any one of the preceding claims 2 to 5, wherein the capability information received for the ML-based application comprises: Information representing goals and / or constraints for performing the function, wherein the goals and / or constraints are defined based on at least one network metric.

7. The apparatus of any preceding claim, wherein the capability information received for an ML-based application comprises: Information representing configuration parameters of an ML model implemented by the ML-based application.

8. The apparatus according to any one of the preceding claims, wherein the method comprises: Mapping information is stored in the intent logic library, the mapping information being used for the mapping performed between the ILU and the at least one corresponding ML-based application.

9. The apparatus according to any one of the preceding claims, wherein the method comprises: receiving a request including intent information indicative of an intent to be implemented with respect to at least one communication network; identifying, based on capability information stored in the intent logic repository, one or more ILUs configured to initiate execution of one or more corresponding ML-based applications, the one or more corresponding ML-based applications being adapted to implement intents for the at least one communication network; The identified one or more ILUs are executed.

10. The apparatus of any one of the preceding claims, wherein the method comprises: The one or more ML-based applications are configured to map with the identified one or more ILUs.

11. An apparatus comprising means for performing a method comprising: receiving a request for capability information from an intent logic execution platform; In response to the request, sending capability information indicating the capability of one or more machine learning (ML) based applications; Executing one or more machine learning (ML) based applications to implement an intent, wherein the execution is initiated by at least one intent logic unit mapped to the one or more ML based applications.

12. A method comprising: sending a request for capability information, wherein the request is sent to at least one of: an ML-based application, an ML application coordinator, and an ML capability producer; Receiving capability information indicating capabilities of one or more machine learning (ML) based applications; generating one or more intent logic units (ILUs), wherein the ILUs are mapped with at least one corresponding ML-based application, wherein the ILUs are configured to initiate execution of the at least one corresponding ML-based application mapped with the ILUs; The one or more ILUs are stored in an intent logic library, wherein the ILUs are stored in association with capability information derived for the ILU from capability information received for the at least one corresponding ML-based application mapped with the ILU.

13. A method comprising: receiving a request for capability information from an intent logic execution platform; In response to the request, sending capability information indicating the capability of one or more machine learning (ML) based applications; Executing one or more machine learning (ML) based applications to implement an intent, wherein the execution is initiated by at least one intent logic unit mapped to the one or more ML based applications.

14. A non-transitory computer-readable medium comprising program instructions stored thereon for performing at least the following: sending a request for capability information, wherein the request is sent to at least one of: an ML-based application, an ML application coordinator, and an ML capability producer; Receiving capability information indicating capabilities of one or more machine learning (ML) based applications; generating one or more intent logic units (ILUs), wherein the ILUs are mapped with at least one corresponding ML-based application, wherein the ILUs are configured to initiate execution of the at least one corresponding ML-based application mapped with the ILUs; The one or more ILUs are stored in an intent logic library, wherein the ILUs are stored in association with capability information derived for the ILU from capability information received for the at least one corresponding ML-based application mapped with the ILU.

15. A non-transitory computer-readable medium comprising program instructions stored thereon for performing at least the following: receiving a request for capability information from an intent logic execution platform; In response to the request, sending capability information indicating the capability of one or more machine learning (ML) based applications; Executing one or more machine learning (ML) based applications to implement an intent, wherein the execution is initiated by at least one intent logic unit mapped to the one or more ML based applications.

Images