Network and method

The implementation of state and time management entities within 6G networks addresses the inefficiencies in intent handler management, optimizing resource allocation and reducing latency, thereby enhancing the performance and hyper-connectivity of 6G systems.

GB2629759APending Publication Date: 2024-11-13SAMSUNG ELECTRONICS CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
GB2023005073
Authority / Receiving Office
GB · GB
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-04-05
Publication Date
2024-11-13

AI Technical Summary

Technical Problem

Current 6G network management systems lack effective state and time management for intent handlers, which are crucial for efficient data processing and resource allocation, especially during observation periods, leading to inefficiencies and increased operational complexity.

Method used

The introduction of a State Management Entity to store and expose intent handler states via an API, a Time Management Entity to schedule and update state timelines, and a Time Estimator Entity to estimate required processing times using AI modules and algorithms, ensuring efficient allocation of resources and observation periods.

Benefits of technology

This solution enables efficient management of intent handlers in 6G networks, reducing latency and operational complexity by optimizing state and time allocation, thereby enhancing the overall performance and hyper-connectivity of 6G systems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000000_0000_ABST
    Figure 00000000_0000_ABST
Patent Text Reader

Abstract

A telecommunications network, for example a 6G network, comprises a state management entity configured to store respective states of intent handlers for respective intents and / or expectations in one or more tables, provide the stored respective states of the intent handlers via an application programming interface, API such as providing the one or more tables via the API; and trigger a subsequent state following a current state. A time management entity is configured to time -schedule respective states of intent handlers and update time schedules of the states. A time estimator comprising an AI module or algorithm is configured to estimate required times for states of intent handlers by analysing actual start times of states and / or by measuring expectations and provide observation times for an intent report.
Need to check novelty before this filing date? Find Prior Art

Description

Field The present invention relates to telecommunications networks, for example 5G and / or 6G networks. Background to the invention 6G communication systems are expected to provide hyper-connectivity as well as the experience for both consumers and industries. By combining AI / ML algorithms, the 6G networks will be able to collect and leverage a huge amount of data associated with billions of connected machines as well as human needs. 6G systems will improve and enable access to information, resources and social services without constraints of time or physical location. However, management of 6G networks will be very complex due to the heterogeneity of data. Therefore, there is a need to automatize the management. Zero-Touch network management is emerged for beyond 5G networks and currently being standardized in 3GPP SA5. Intent based automation is key to enable zero-touch network management and currently being standardized in 3GPP TS 28312 v17.2.0 and 3GPP TR 28.912 v1.1.0. Intent is defined as the formal specification of all expectations including requirements, goals, and constraints given to a technical system. How to express an intent is modelled in 3GPP TS 28312 V17.2.0. Summary of the Invention A first aspect provides a telecommunications network, for example a 6G network, comprising a state management entity configured to: store respective states of intent handlers for respective intents and / or expectations, for example in one or more tables; optionally, provide the stored respective states of the intent handlers for the respective intents and / or expectations, for example via an application programming interface, API such as providing the one or more tables via the API; and optionally, trigger a subsequent state, for example following a current state such as at the end of the current state. A second aspect provides a telecommunications network, for example a 6G network, comprising a time management entity configured to: time-schedule respective states of intent handlers; and optionally, update respective time-schedules of the states. A third aspect provides a telecommunications network, for example a 6G network, comprising a time estimator entity, optionally comprising an Al module and / or algorithm, wherein the time estimator entity is configured to: estimate respective required times for states of intent handlers for respective intents and / or expectations, for example by analysing respective actual start times of the respective states and / or by measuring respective expectations; and optionally, provide respective observationTimes, for example for an intent report. A fourth aspect provides a method of managing a telecommunications network, for example a 6G network, the method comprising: storing respective states of intent handlers for respective intents and / or expectations, for example in one or more tables; optionally, providing the stored respective states of the intent handlers for the respective intents and / or expectations, for example via an application programming interface, API such as providing the one or more tables via the API; and optionally, triggering a subsequent state, for example following a current state such as at the end of the current state. A fifth aspect provides a method of managing a telecommunications network, for example a 6G network, the method comprising: time-scheduling respective states of intent handlers; and optionally, updating respective time-schedules of the states. A sixth aspect provides a method of managing a telecommunications network, for example a 6G network, the method comprising: estimating respective required times for states of intent handlers for respective intents and / or expectations, for example by analysing respective actual start times of the respective states and / or by measuring respective expectations; and optionally, providing respective observationTimes, for example for an intent report. Detailed Description of the Invention According to the present invention there is provided a telecommunications network, as set forth in the appended claims. Also provided is a method. Other features of the invention will be apparent from the dependent claims, and the description that follows. In 3GPP TR 28.912 v1.1.0, how to report the intent is studied, where “observationTime" is proposed. The “observationTime” represents the MnS consumer’s requirements for the observation period of performance value for corresponding Expectation Targets. The performance value for corresponding Expectation Targets will be observed from the start of each observation period, then at the end of each observation period, the value will be derived and configured. However, it is not specified how to manage the states of an intent handler and how much time of the observationTime the intent handler needs to allocate for each state of intent processing, i.e. data grounding. Therefore, time and / or state management of the intent handler remains to be addressed. Particularly, the aspects described herein address time and / or state management of the intent handler. The first aspect provides a telecommunications network, for example a 6G network, comprising a state management entity configured to: store respective states of intent handlers for respective intents and / or expectations, for example in one or more tables; optionally, provide the stored respective states of the intent handlers for the respective intents and / or expectations, for example via an application programming interface, API such as providing the one or more tables via the API; and optionally, trigger a subsequent state, for example following a current state such as at the end of the current state. The second aspect provides a telecommunications network, for example a 6G network, comprising a time management entity configured to: time-schedule respective states of intent handlers; and optionally, update respective time-schedules of the states. The second aspect may include any subject matter described with respect to the first aspect mutatis mutandis. The third aspect provides a telecommunications network, for example a 6G network, comprising a time estimator entity, optionally comprising an Al module and / or algorithm, wherein the time estimator entity is configured to: estimate respective required times for states of intent handlers for respective intents and / or expectations, for example by analysing respective actual start times of the respective states and / or by measuring respective expectations; and optionally, provide respective observationTimes, for example for an intent report. The third aspect may include any subject matter described with respect to the first aspect and / or the second aspect mutatis mutandis. In one example, the Al module and / or algorithm is adapted for setting times for respective states. In one example, the Al module and / or algorithm is adapted for analysing time in respective states and optionally, updating time allocation for respective states. In one example, the Al module and / or algorithm is adapted for updating respective times allocated for the states, for example based on measurement of actual expectation. In one example, the Al module and / or algorithm is adapted for scheduling respective intents based on respective stored states, for example provided via an application programming interface, API. The fourth aspect provides a method of managing a telecommunications network, for example a 6G network, the method comprising: storing respective states of intent handlers for respective intents and / or expectations, for example in one or more tables; optionally, providing the stored respective states of the intent handlers for the respective intents and / or expectations, for example via an application programming interface, API such as providing the one or more tables via the API; and optionally, triggering a subsequent state, for example following a current state such as at the end of the current state. The fifth aspect provides a method of managing a telecommunications network, for example a 6G network, the method comprising: time-scheduling respective states of intent handlers; and optionally, updating respective time-schedules of the states. The fifth aspect may include any subject matter described with respect to the fourth aspect mutatis mutandis. The sixth aspect provides a method of managing a telecommunications network, for example a 6G network, the method comprising: estimating respective required times for states of intent handlers for respective intents and / or expectations, for example by analysing respective actual start times of the respective states and / or by measuring respective expectations; and optionally, providing respective observationTimes, for example for an intent report. The sixth aspect may include any subject matter described with respect to the fourth aspect and / or the fifth aspect mutatis mutandis. In one example, the method comprises setting times for respective states. In one example, the method comprises analysing time in respective states and optionally, updating time allocation for respective states. In one example, the method comprises updating respective times allocated for the states, for example based on measurement of actual expectation. In one example, the method comprises scheduling respective intents based on respective stored states, for example provided via an application programming interface, API. Definitions Throughout this specification, the term “comprising” or “comprises” means including the component(s) specified but not to the exclusion of the presence of other components. The term “consisting essentially of’ or “consists essentially of’ means including the components specified but excluding other components except for materials present as impurities, unavoidable materials present as a result of processes used to provide the components, and components added for a purpose other than achieving the technical effect of the invention, such as colourants, and the like. The term “consisting of’ or “consists of’ means including the components specified but excluding other components. Whenever appropriate, depending upon the context, the use of the term “comprises” or “comprising” may also be taken to include the meaning “consists essentially of’ or “consisting essentially of’, and also may also be taken to include the meaning “consists of’ or “consisting of’. The optional features set out herein may be used either individually or in combination with each other where appropriate and particularly in the combinations as set out in the accompanying claims. The optional features for each aspect or exemplary embodiment of the invention, as set out herein are also applicable to all other aspects or exemplary embodiments of the invention, where appropriate. In other words, the skilled person reading this specification should consider the optional features for each aspect or exemplary embodiment of the invention as interchangeable and combinable between different aspects and exemplary embodiments. Brief description of the drawings For a better understanding of the invention, and to show how exemplary embodiments of the same may be brought into effect, reference will be made, by way of example only, to the accompanying diagrammatic Figures, in which: Figure 1: Functional Architecture of State of Art Cognitive Layer [1]; Figure 2: State of Art Intent Manager Framework [2]; Figure 3: Basic Diagram of Reinforcement Learning. RL State is the input to the agents, where it represents the current information of the environment. Action is output from the agent based on the State of the environment, which changes the current State. Reward is the feedback (positive or negative) from the environment; Figure 4: Adding Time and State Management in the state-of-art Intent Handler - proposed entities are shown in green colour; Figure 5: Detailed View of Time and State Management in the state-of-art Intent Handler, proposed entities are shown in green colour; Figure 6: Example Sequence Diagram; Figure 7: Functional View of State Management Entity - Proposed functions and interfaces are shown in green colour; Figure 8: Example Sequence Diagram of State Management Entity; Figure 9. Example relational diagram of the intent handler states; Figure 10: Functional View of Time Management Entity - Proposed functions and interfaces are shown in green colour; Figure 11: Example sequence diagram of Time Management Entity; Figure 12: Functional view of Time Estimator Entity - Proposed functions and interfaces are shown in green colour; Figure 13: Example Sequence Diagram of Time Estimator Entity; Figure 14: Example of Al module time estimator; and Figure 15: Intent Scheduler - Proposed functions and interfaces are shown in green colour. Detailed Description of the Drawings Although a preferred embodiment has been shown and described, it will be appreciated by those skilled in the art that various changes and modifications might be made without departing from the scope of the invention, as defined in the appended claims and as described above. The cognitive layer of [1] is then implemented in a prototype of [2], where cognitive layer forms the intent manager framework given in Figure 2 and, together with cognitive agents (i.e. data grounding agent, proposal agent, prediction agent, evaluation agent, actuator agent), it enables the intent handlers. As shown in the prototype of [2], the intent handler then processes that data and finds the best action to take in order to meet the expectation targets [2]. In the state-of-art intent handler, also known as intent manager framework [2], there is no mechanism for time management of intent handler(s). Developments in AI / ML introduce novel data-driven solutions where multiple aspects of the mobile network optimisation are performed by Al agents that continuously improve the network parameters. Reinforcement Learning (RL) is an example of Al methods, which considered a promising approach for many problems in mobile network optimisation. The basic idea of the RL agent is to learn through trial-and-error and getting feedback from its actions. This feedback can be either a “reward” or a “penalty”, depending on the action it performed. In 3GPP TR 28.912 v1.1.0, how to report the intent is studied, where “observationTime" is proposed. The “observationTime” represents the MnS consumer’s requirements for the observation period of performance value for corresponding Expectation Targets. However, the issue how intent handler needs to manage its time during the “observationTime” remains to be addressed. In this invention, we propose method and apparatus for time and state management of intent handlers in order to assure the intent handler to spend its time in order to efficiently manage and execute the required states during the specified observationTime for each intent. The current state-of-art intent handlers do not have state management entity which manages the states and exposes the states through an API. In this invention, we propose such entity to keep record of each state for intent handlers and expose it for better observability. The current state-of-art intent handlers do not have time management entity, we propose such entity and algorithms and / or Al modules that facilitates to manage the timing of each state and / or schedule the states and / or expose the time required for the intent processing as suggested observationTime. This invention will enable the time and state management for intent handlers of 6G systems which will bring the hyper-connected experience to every corner of life and this will require a “systems of systems” approach which is a set of dedicated systems that offers their resources and capabilities together to create a new, more complex system. As intent-based networking simplifies the management by only expressing the expectations, it will enable Al-driven 6G network management and reduce the OPEX significantly. In each domain - not limited to 3GPP entities - it is expected to deploy an intent handler to process the intents of the consumer which could be the end user (service operators, user equipment, factory machines etc.) as well as the other entities forming the 6G network (i.e. Core Network, RAN network, Orchestrator, Edge domain etc.). The intent owner (also known as the MnS consumer) only focuses on their expectations rather than how to reach them which is the responsibility of an intent handler. Therefore, the intent owner does not care about the heterogeneity of the data that will help them reach their expectations which are delegated as goals to the intent handler. As the latency requirements will be stringent in 6G networks, new technologies will be required to use the time efficiently for the intent handlers to reach their goals. Therefore, this invention tackles at the problem of how to efficiently manage the time and state for intent based automation of 6G and expose it for further improvement of the 6G network management. 3.3 Drawings describing the invention (block diagrams, flow charts, time diagrams) The proposed entities in this invention are shown in Figure 4 and 5. The claims of this inventions are as follows: • Independent claim includes State Management Entity where it keeps the states of intent handlers in table(s) for each intent and / or expectation, expose the table(s) through a specific API, and triggers the next state at the end of each state. • Independent claim includes Time Management Entity which time-schedules for each state of the intent handler and updates the time-schedules for each state • Independent claim includes a Time Estimator which estimates the required time for each state for each intent and / or expectation by analysing the actual start time of each state, and / or measurement of expectation(s), and writes a suggested observationTime in the knowledge which can be encapsulated in the intent report. • Dependent claim includes an Al module and / or algorithm in the Time Estimator Entity that initially set times to spend for each state. • Dependent claim includes an Al module and / or algorithm in the Time Estimator Entity that analyses time spent in each state and updates time allocation for each state. • Dependent claim includes an Al module and / or algorithm in the Time Estimator Entity to update the times allocated for each state based on current measurement of actual expectation. • Dependent claim includes an Al module and / or algorithm for the intent owner to schedule intents based on the exposed state tables through a specific API. An example sequence diagram of the proposed method and apparatus is given in Figure 6. As shown in the example sequence diagram, the intent owner registers their intent through the intent handler’s user interface. As the intent is registered in the knowledge base, the state management entity is triggered to create state tables for the intent and / or expectations. The state management entity then requests the time management entity to initially schedule each state. The time management entity requests estimated time to spend for each state from the time estimator entity and upon receiving the estimates, it schedules each state and updates the state table in the state management entity. The state management entity triggers each state and register the start time for each state as they are started. Then the state management entity requests the time management entity to update the time schedules with sending the table with actual start times. The time management entity requests the time estimator entity to update the estimated time to spend for each state, and upon receiving the estimates, the time management entity to update the state table with new time schedules. In the meantime, the time estimator entity continuously grounds the measurement of the target expectation and register the time when the target expectation is FULFILLED and writes this time as the recommended observation time to the knowledge. In addition, the state management entity exposes the state table through an API when it is requested. This table then can be used to schedule the intents to be placed to the intent handler by the intent owner as well as to improve efficiency of 6G networks. 3.3.1 State Management Entity The state management entity is responsible for keeping the states of intent handlers for each intent, and triggers the next state at the end of each state. When an intent is submitted, it is automatically registered in the knowledge. From its registration to knowledge, the states are continuously performed until the intent is removed by the MnS consumer (or intent may be expired as stated in Intent Context). These states are managed by the state management entity keeping each state status in a table which is created upon intent registration in the knowledge. State Management entity includes following components as shown in Figure 7: • State Controller Function which is responsible for processing, writing and exposing state tables for each intent and / or expectation(s) • Table Creator Function which is responsible for creating a state table for each intent and / or expectation(s) •     State Trigger Function which is responsible for triggering a state through reasoning •     State Database which is responsible for storing state tables. An example sequence diagram of State Management Entity is provided in Figure 8. State Controller Function subscribes knowledge base for new registered intents and upon registration, it sends a request to Table Creator Function to create table(s) for the intent and / or expectation(s). These tables should include each state of intent handler which are described as follows and given in Figure 9 with their order: • Registration State: This is the state when the intent is registered. It can be triggered as an initial step through intent interface or can be triggered after reporting state, escalation state and coordination state as a result of updating the intent and / or placing a new intent. • Knowledge State: Knowledge state is when there is a knowledge and / or knowledge graph inserted and / or updated and / or removed in the knowledge base. • Reasoning State: Reasoning is the state when using inference to decide the following state (i.e. Data State, Proposal State, Prediction State, Evaluation State, Actuation State, Reporting State). Reasoning may include algorithms and / or policies and / or Al modules which are deployed for specific reason (i.e. selecting which data to ground, if there is an issue to trigger proposal state, etc.) • Data State: Data state is the state when raw data which is relevant to the expectation is being grounded from managed entity. Please note that grounding the actual expectation measurement (i.e. latency of a UE) is continuous throughout every state of an intent handler, started at the registration state until the intent is removed by the MnS consumer. • Proposal State: Proposal state is the state when an action is proposed. It is triggered after the Reasoner finds an issue with an intent expectation. • Prediction State: Prediction state is the state to provide predictions for a proposed action, triggered by the Reasoner. • Evaluation State: Evaluation state is the state to evaluate the proposal and prediction for an issue, triggered by the Reasoner. • Actuation State: Actuation state is the state when an action is executed in the managed entity • Reporting State: Reporting state is the state when a report is being sent to the intent owner / consumer. • Training State: Training state is the state when any Al module deployed in the intent handler (i.e. in the Reasoner, Proposal agent, Prediction agent etc.) • Escalation State: Escalation state is the state when the intent is decided that it will not be fulfilled and escalated back to the MnS consumer. Training State and Escalation State are shown with dashed lines in Figure 9 to indicate that these states may not be triggered in each cycle of intent handling. For example, Training State may be triggered when the Reasoner decides that some Al modules are not working as expected or Escalation State may be triggered when the Reasoner decides there is no way that intent handler can fulfil the submitted intent and escalates it to the MnS consumer. Reporting state is always triggered at the end of the observationTime specified by the MnS Consumer, regardless of what state the intent handler is. 5 After the table is created and sent to State Controller Function from Table Creation Function, the State Controller Function requests Time Management Entity to initially schedule each state. Upon scheduling is provided, the State Controller Function associates the table with intent ID, expectation ID and their observationTime and writes the table in the State Database. 10 Then, the State Controller Function requests the State Trigger Function to trigger the states based on scheduled times. The State Trigger Function sends a request to the Reasoner to facilitate the next state. Upon successful response from the Reasoner, the State Trigger Function sends the actual start time of each state to the State Controller Function. The State Controller Function then writes these times to the table(s). Intent ID Expectation ID observationTime State Previous State Following State Estimated Time to Spend Scheduled Time Actual Start Time Inti Exp1 0:0:30,00 Reasoning State Knowledge State Data State 0.0:1,00 22-02-11 16:35:47,985 +0000 22-02-11 16:35:47,992 +0000 Inti Exp1 0:0:30,00 Data State Reasoning State Knowledge State 0.0:5,50 22-02-11 16:35:48,985 +0000 22-02-11 16:35:48,002 +0000 Inti Exp1 0:0:30,00 Knowledge State Data State Reasoning State 0.0:0,44 22-02-11 16:35:54,485 +0000 22-02-11 16:35:53,003 +0000 Inti Exp1 0:0:30,00 Reasoning State Knowledge State Proposal State 0.0:2,25 22-02-11 16:35:54,925 +0000 22-02-11 16:35:53,921 +0000 Inti Exp1 0:0:30,00 Proposal State Reasoning State Knowledge State 0.0:6,50 22-02-11 16:35:57,175 +0000 22-02-11 16:35:56,953 +0000 Inti Exp1 0:0:30,00 Knowledge State Proposal State Reasoning State 0.0:0,44 22-02-11 16:36:03,675 +0000 22-02-11 16:36:00,985 +0000 Inti Exp1 0:0:30,00 Reasoning State Knowledge State Prediction State 0.0:1,25 22-02-11 16:36:04,115 +0000 22-02-11 16:36:06,985 +0000 Inti Exp1 0:0:30,00 Prediction State Reasoning State Knowledge State 0,0:4,50 22-02-11 16:36:05,335 +0000 22-02-11 16:36:07,985 +0000 Inti Exp1 0:0:30,00 Knowledge State Prediction State Reasoning State 0.0:0,44 22-02-11 16:36:09,835 +0000 22-02-11 16:36:10,985 +0000 Inti Exp1 0:0:30,00 Reasoning State Knowledge State Evaluation State 0.0:0,44 22-02-11 16:36:10,385 +0000 22-02-11 16:36:11,985 +0000 Inti Exp1 0:0:30,00 Evaluation State Reasoning State Knowledge State 0.0:2,00 22-02-11 16:36:10,785 +0000 22-02-11 16:36:12,985 +0000 Int1 Exp1 0:0:30,00 Knowledge State Evaluation State Reasoning State 0.0:0,44 22-02-11 16:36:12,125 +0000 22-02-11 16:36:24,985 +0000 Inti Exp1 0:0:30,00 Reasoning State Knowledge State Actuating State 0.0:0,44 22-02-11 16:36:12,665 +0000 22-02-11 16:36:25,085 +0000 Inti Exp1 0:0:30,00 Actuating State Reasoning State Knowledge State 0.0:2,00 22-02-11 16:36:13,105 +0000 22-02-11 16:36:25,185 +0000 Inti Exp1 0:0:30,00 Knowledge State Actuating State Reasoning State 0.0:0,44 22-02-11 16:36:15,105 +0000 22-02-11 16:36:25,985 +0000 Inti Exp1 0:0:30,00 Reasoning State Knowledge State Reporting State 0.0:0,44 22-02-11 16:36:16,545 +0000 22-02-11 16:36:26,085 +0000 Exp1 0:0:30,00 Reporting State Reasoning State Knowledge State 0.0:1,00 22-02-11 16:36:16,985 +0000 22-02-11 16:36:26,885 +0000 Table 1: Example Table of States for a given intent and expectation at a given time X. Coloring is explained in Section 3.3.2.1 Among the states given in Table 1, the most time-consuming states would be Data State as the initial data is collected and processed from the managed entity over a chosen period of time, and Reasoning State, Proposal State, Prediction State, Evaluation State as they may need to run algorithms and / or Al modules. Upon completing the table(s) at the end of the observationTime, the State Controller Function queries the knowledge base whether the intent is still registered and not removed by the MnS Consumer. Upon confirmation, the State Controller Function requests the Time Management Entity to update the scheduled times. Upon the updates are received, the State Controller Function updates the tables in the State Database and continue with registration of the actual times as specified in Steps 8,9,10,11,12 and 13 in Figure 8. The tables are created, updated and stored inside the state management entity of the intent handler and exposed by the State Controller Function via an API to MnS consumers and / or any software agents, end users (i.e. service operators, user equipment, factory machines etc.) as well as any other managed network entity intent handler (i.e. Core Network, RAN network, Orchestrator, etc.) 3.3.2 Time Management Entity The roles of the time management entity are as follows: •     Scheduling of each state for each intent and / or expectation as shown in Table 1 •     Continuously updating scheduled time for each state as the actual times are reported by the state management entity and new time estimates are reported by the Time Estimator Entity. • If estimated time for a state is zero, then time management entity does not write any schedule time for that state, so that state is not triggered by the state management entity. The Time Management Entity includes the following components and shown in Figure 10: Time Controller Function • Time Scheduler Function An example sequence diagram is provided in Figure 11. Upon receiving the request from the State Management Entity, the Time Controller Function sends a request to Time Scheduler Function to initially schedule time for each state given in the table and sent within the request. Then Time Scheduler Function sends a request to Time Estimator Entity to receive estimated time to spend for each state. Upon receiving the estimated time, the time scheduler function creates initial time schedules leveraging the current time stamp and sends them to Time Controller Function. Time Controller Function then sends the table with initial time schedules to the State Management Entity. Then, the State Management Entity may request to update the time schedules based on the actual start times of each state. In this case, the request includes the table with actual start times. The Time Controller Function then sends a request to Time Scheduler Function to update time schedules. Time Scheduler Function sends a request to Time Estimator Entity to update estimated time to spend for each state. Upon receiving the updated estimations, the Time Scheduler Function responds new time schedules for each state and then Time Controller Function submits updated schedules to State Management Entity. 3.3.3 Time Estimator Entity The Time Estimator Entity include the following components as shown in Figure 12: • Estimator Controller Function which is responsible for processing any estimation requests and sending the estimated values. • Initial Estimator Function which is responsible for initially estimating time to be spent for each state using data from knowledge (Section 3.3.3.1) • Estimate Updater Function which is responsible for estimating time to be spent for each state using their actual start times (Section 3.3.3.2) as well as updating the time estimates for each state observing the measurement of the actual expectation (Section 3.3.3.3) • Required Time Calculator Function which is responsible for writing the suggested observationTime required for each expectation and intent (Section 3.3.3.4) For estimating the times for each state, the Time Estimator Entity may leverage the Al module which is described further in Section 3.3.3.5. An example sequence diagram for Time Estimator Entity is given in Figure 13. As given in Figure 13, the Estimator Controller Function receives a request from the Time Management Entity to provide initial estimated times to spend for each state. This request includes the state tables without any previously registered time schedules and actual start times. The Estimator Controller Function understands this request is for initial time estimates and sends it to the Initial Estimator Function. The Initial Estimator Function leverages the method described in Section 3.3.3.1 to find initial times to spend for each state and then writes the estimated time to spend in the table and send it back to the Estimator Controller Function. If the Estimator Controller Function receives the request from the Time Management Entity to provide estimated times to spend for each state and the request includes the state tables with previous time schedules and actual start times, then the Estimator Controller Function sends this request to Estimate Updater Function. The Estimate Updater Function leverages the method described in Section 3.3.3.2 to find the required times to spend for each state and writes the estimated time to spend in the table and send it back to the Estimator Controller Function. The Estimate Updater Function continuously observe the measurement of the actual expectation(s) of the intent(s) (i.e. latency value of a UE). Based on the observation described in Section 3.3.3.3, the Estimate Updater Function may update the estimate time to spend for the states as 0, if the expectation(s) are FULFILLED. As Estimate Updater Function updates the estimated times, it sends these times to the Estimate Controller Function which then sends to the Time Management Entity. The Estimate Updater Function also sends the time estimates to the Required Time Calculator Function. The Required Time Calculator Function calculates and writes the suggested observationTime in the knowledge base as described in Section 3.3.3.4. The Estimate Updater Function may deploy Al module to perform estimations as discussed in Section 3.3.3.5. 3.3.3.1 Initial Time Estimation of Each State The Initial Estimator Function may use an algorithm and / or domain knowledge to schedule the states. For example, from the domain knowledge, it may be observed that the states that are highlighted in red in Table 1 are the ones that are time-consuming. Data State requires the initial data is grounded from the managed entity over a chosen period of time, and Reasoning State, Proposal State, Prediction State, Evaluation State as they may need to run algorithms and / or Al modules. So initial scheduling may include allocation more time to these states whereas allocating standard observed processing time to the remaining states. Alternatively, the Al module may be deployed to decide the optimum initial schedule time based on computational complexity of each state. The Initial Estimator Function may also have the knowledge of required observationTime for expectations of the submitted intent. In this case, if the observationTime specified by the MnS consumer is less than required observationTime, in this case the Management Entity does not schedule the states in Table 1 except the reporting state which is triggered at the observationTime and reports that the expectations are in NOTFULFILLED state. Reporting state is always scheduled at the observationTime which is specified by the MnS consumer and states when the MnS consumer expects an intent report from the intent handler. 3.3.3.2 Updating the state schedules based on actual start times Whenever a state is completed, next state is triggered and its start time is registered in the table. Actual start time may be different from initial schedule time due to many reasons such as: • Erroneous initial scheduling • Health problems of intent handlers (i.e. not enough resources, too many intents to handle etc.) An algorithm and / or Al module may be deployed to leverage initial scheduled times, actual start times and as well as data on the system gathered from the knowledge (i.e. datacenter usage of intent handler, resource plans etc.) in order to decide the new schedule times for each state of the intent handler for each expectation and / or intent. 3.3.3.3 Updating the state schedules based on the measurement of the actual expectation The Estimate Updater Function continuously observe the measurement of the actual expectation(s) of the intent(s) (i.e. latency value of a UE). Whenever the expectation is measured within the target value over a specific time, the estimated time to spend for the states associated with those expectations will be zero. The Time Management Entity does not schedule those states of which estimated time to spend is zero. The State Management Entity also continuously observe the measurement of the actual expectation(s) of the intent(s). Whenever the expectation is measured within the target value over a specific time, the State Management Entity deletes the scheduled times for the remaining states until the reporting state and these remaining states are not triggered and the reporting state is triggered at the end of the observationTime. As the remaining states are not performed, these resources may be allocated for the states of the other expectation(s) and / or intent(s) 3.3.3.4 Writing the suggested observationTime to the Knowledge Upon the Estimate Updater Function has finished updating the time estimates, it sends them to Required Time Calculator Function which sums all the estimated time durations to spend for each state, and this sum is denoted by ¢. This value is kept with a time-stamp of the time of calculation, t1, until the new values are sent. As stated in the previous section, the Estimate Updater Function continuously observe the measurement of the actual expectation(s) of the intent(s). Upon FULFILLMENT of intent and / or expectation(s) over a specific time, the Estimate Updater Function sends the timestamp, t2, when the FULFILLMENT is decided to the Required Time Calculator Function which then calculates the suggested observationTime as ¢+( t2- ti ) and writes this value to knowledge base. This value gives a good estimation of observationTime as: • It covers the time between execution of an action and observation of its effect on target KPI • It does not get effected by false FULFILLMENT states which can be obtained before executing an action due to the noise in the system 3.3.3.5 Al module for Time Estimator The Time Estimator Entity may include an Al module and / or algorithm. The Al module can be modelled as RL problem, where it predicts the estimating time to be spent for each state. The estimation time evaluated by the Time Management by comparing the predicated time with the actual time and send the feedback (positive or negative) to the Al module (see figure 9). 3.3.4 Scheduling Intents Using the Table of States As shown in the Example Sequence Diagram of Figure 6, the MnS consumer (also known as the intent owner) may request the states table from the state management entity through a specific API. The MnS consumer may then collect other relevant data (e.g. available resources) together with the table of states and run algorithms and / or Al modules to schedule all intents that are to be registered in the intent handler in order not to cause any increase in system overhead. 3.4 Alternatives implementations of the invention 3.5 Advantages & Novel features of the invention The current state-of-art intent handlers do not have state management entity which manages the states and exposes the states through an API. In this invention, we propose such entity to keep record of each state for intent handlers and expose it for better observability. The current state-of-art intent handlers do not have time management entity, we propose such entity and algorithms and / or Al modules that facilitates to manage the scheduling the states. The current state-of-art intent handlers do not have time estimator entity, we propose such entity and algorithms and / or Al modules that estimates time duration to spend for each state as well as calculates suggested observationTime. The current state-of-art intent handlers do not have intent schedules, we propose intent scheduler function to schedule intents based on the state tables exposed by the state management entity. <Advantages over existing technology> The advantages of the proposed method and apparatus are as follows: Scheduling each state of intent handler and finding required time for each state. Managing the time and thus resources more efficiently for the intent handlers. Exposing the states and their timing through an API for better observability of intent handlers. Suggesting observationTime that can be delegated to the MnS Consumer. References [1] J Nielmoller et al. Cognitive Processes for adaptive intent based networking Available at intent-b^ [2] A. C. Baktir et al., "Intent-based cognitive closed-loop management with built-in conflict handling," 2022 IEEE 8th International Conference on Network Softwarization (NetSoft), Milan, Italy, 2022, pp. 73-78, doi: 10.1109 / NetSoft54395.2022.9844074. [3] 3GPPTR 28.912 v1.1.0 [4] S5-232455 Notes At least some of the example embodiments described herein may be constructed, partially or wholly, using dedicated special-purpose hardware. Terms such as ‘component’, ‘module’ or ‘unit’ used herein may include, but are not limited to, a hardware device, such as circuitry in the form of discrete or integrated components, a Field Programmable Gate Array (FPGA) or Application Specific Integrated Circuit (ASIC), which performs certain tasks or provides the associated functionality. In some embodiments, the described elements may be configured to reside on a tangible, persistent, addressable storage medium and may be configured to execute on one or more processors. These functional elements may in some embodiments include, by way of example, components, such as software components, object-oriented software components, class components and task components, processes, functions, attributes, procedures, subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays, and variables. Although the example embodiments have been described with reference to the components, modules and units discussed herein, such functional elements may be combined into fewer elements or separated into additional elements. Various combinations of optional features have been described herein, and it will be appreciated that described features may be combined in any suitable combination. In particular, the features of any one example embodiment may be combined with features of any other embodiment, as appropriate, except where such combinations are mutually exclusive. Throughout this specification, the term “comprising” or “comprises” means including the component(s) specified but not to the exclusion of the presence of others. Attention is directed to all papers and documents which are filed concurrently with or previous to this specification in connection with this application and which are open to public inspection with this specification, and the contents of all such papers and documents are incorporated herein by reference. All of the features disclosed in this specification (including any accompanying claims, abstract 5 and drawings), and / or all of the steps of any method or process so disclosed, may be combined in any combination, except combinations where at least some of such features and / or steps are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and 10 drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise. Thus, unless expressly stated otherwise, each feature disclosed is one example only of a generic series of equivalent or similar features. The invention is not restricted to the details of the foregoing embodiment(s). The invention 15 extends to any novel one, or any novel combination, of the features disclosed in this specification (including any accompanying claims, abstract and drawings), or to any novel one, or any novel combination, of the steps of any method or process so disclosed.

Claims

1. A telecommunications network, for example a 6G network, comprising a state management entity configured to:store respective states of intent handlers for respective intents and / or expectations, for example in one or more tables;optionally, provide the stored respective states of the intent handlers for the respective intents and / or expectations, for example via an application programming interface, API such as providing the one or more tables via the API; andoptionally, trigger a subsequent state, for example following a current state such as at the end of the current state.

2. A telecommunications network, for example a 6G network, comprising a time management entity configured to:time-schedule respective states of intent handlers; and optionally, update respective time-schedules of the states.

3. A telecommunications network, for example a 6G network, comprising a time estimator entity, optionally comprising an Al module and / or algorithm, wherein the time estimator entity is configured to:estimate respective required times for states of intent handlers for respective intents and / or expectations, for example by analysing respective actual start times of the respective states and / or by measuring respective expectations; andoptionally, provide respective observationTimes, for example for an intent report.

4. The network according to claim 3, wherein the Al module and / or algorithm is adapted for setting times for respective states.

5. The network according to any of claims 3 to 4, wherein the Al module and / or algorithm is adapted for analysing time in respective states and optionally, updating time allocation for respective states.

6. The network according to any of claims 3 to 5, wherein the Al module and / or algorithm is adapted for updating respective times allocated for the states, for example based on measurement of actual expectation.

7. The network according to any of claims 3 to 5, wherein the Al module and / or algorithm is adapted for scheduling respective intents based on respective stored states, for example provided via an application programming interface, API.

8. A method of managing a telecommunications network, for example a 6G network, the method comprising:storing respective states of intent handlers for respective intents and / or expectations, for example in one or more tables;optionally, providing the stored respective states of the intent handlers for the respective intents and / or expectations, for example via an application programming interface, API such as providing the one or more tables via the API; andoptionally, triggering a subsequent state, for example following a current state such as at the end of the current state.

9. A method of managing a telecommunications network, for example a 6G network, the method comprising:time-scheduling respective states of intent handlers; andoptionally, updating respective time-schedules of the states.

10. A method of managing a telecommunications network, for example a 6G network, the method comprising:estimating respective required times for states of intent handlers for respective intents and / or expectations, for example by analysing respective actual start times of the respective states and / or by measuring respective expectations; andoptionally, providing respective observationTimes, for example for an intent report.

11. The method according to claim 10, comprising setting times for respective states.

12. The method according to any of claims 10 to 11, comprising analysing time in respective states and optionally, updating time allocation for respective states.

13. The method according to any of claims 10 to 12, comprising updating respective times allocated for the states, for example based on measurement of actual expectation.

14. The method according to any of claims 10 to 13, comprising scheduling respective intents based on respective stored states, for example provided via an application programming interface, API.