AI-driven intention and resource utilization optimization method

Abstract

According to one aspect, there is provided a computer-implemented method performed by a first network node in a communication network environment for optimizing transmission parameters of at least one user equipment. The method includes obtaining a first set of at least one intent of at least one user device. The method includes obtaining at least one intent preference value associated with the at least one user device based on the obtained first set of at least one intent. The method includes determining a communication configuration of the at least one user equipment based on the at least one intent preference value using a first AI / ML model. The method includes causing a change to a configuration of the at least one user equipment based on the communication configuration.

Description

Technical Field

[0001] Examples involving the use of artificial intelligence (AI) and machine learning (ML) technologies to optimize intent and resource utilization are disclosed. Background Technology

[0002] Figure 1 The 5G RAN (NG-RAN) architecture is illustrated. The 5G RAN architecture is described in the 3GPP Technical Specification (TS) 38.401 v. 15.8.0. NG-RAN 102 consists of a group of gNBs (104A, 104B) connected to the 5GC (100) via NG interfaces. As specified in 3GPP TS 38.300, v. 17.5.0, NG-RAN 102 can also consist of a group of ng-eNBs, which can be composed of ng-eNB-CUs and one or more ng-eNB-DUs. The ng-eNB-CUs and ng-eNB-DUs are connected via W1 interfaces. gNBs can support FDD, TDD, or dual-mode operation. gNBs can interconnect via Xn interfaces.

[0003] gNBs 104A and 104B can consist of a gNB-CU 106 and one or more gNB-DUs 108A-B. The gNB-CU and gNB-DU are connected via the F1 interface. A gNB-DU is connected to only one gNB-CU. NG, Xn, and F1 are logical interfaces. For NG-RAN, the NG and Xn-C interfaces for a gNB consisting of a gNB-CU and gNB-DU terminate at the gNB-CU. For EN-DC, the S1-U and X2-C interfaces for a gNB consisting of a gNB-CU and gNB-DU terminate at the gNB-CU. The gNB-CU and connected gNB-DU are only visible to other gNBs and the 5GC as a gNB.

[0004] Figure 2 The architecture used to separate gNB-CU-CP and gNB-CU-UP is shown. Figure 2 Including gNB-CU-CP 206, gNB-CU-UP 210 and gNB-DU 208A-B.

[0005] 3GPP RP-201620 defines a research project (SI) "Enhancements for Data Collection in NR and EN-DC". This research project aims to investigate the functional framework for RAN intelligence that further enhances data collection through use cases, examples, etc., and to identify potential standardization impacts on current NG-RAN nodes and interfaces.

[0006] The detailed objectives of this SI are as follows: to study the high-level principles and functional framework of AI-enabled RAN intelligence (e.g., inputs / outputs of AI functions and components optimized for AI-enabled RAN), and to identify the benefits of AI-enabled NG-RAN through possible use cases (e.g., energy saving, load balancing, mobility management, coverage optimization, etc.).

[0007] (a) Study the standardization impact on identified use cases, including: the data that the AI ​​function may need as input, and the data that the AI ​​function may generate as output that is interpretable for multi-vendor support.

[0008] (b) Study the impact of standardization on nodes or functions that receive / provide input / output data in the current NG-RAN architecture.

[0009] (c) Investigate the impact of standardization on network interfaces to facilitate the transfer of input / output data between network nodes or AI functions.

[0010] As part of SI work, a provisional version of 3GPP Technical Report (TR) 37.817 v. 17.0.0 was made available after RAN3#114-bis-e, and the following text is noted.

[0011] AI-enabled RAN intelligence should adopt the following high-level principles:

[0012] The detailed AI / ML algorithms and models for the use cases are implementation-specific and go beyond the scope of RAN3.

[0013] This study focuses on AI / ML functionality and the corresponding input / output types.

[0014] The inputs / outputs and locations of model training and model inference functions should be studied on a case-by-case basis.

[0015] The study focuses on the analysis of the data required for model training functions from the perspective of data collection, while the aspects of how model training functions use inputs to train models go beyond the scope of RAN3.

[0016] The study focuses on the analysis of the data required for the model inference function from the perspective of data collection, while the aspects of how the model inference function uses inputs to derive outputs are beyond the scope of RAN3.

[0017] The placement of AI / ML capabilities within the current RAN architecture depends on deployment and specific use cases.

[0018] If needed, model training and inference functions should be able to request specific information for training or executing AI / ML algorithms and avoid receiving unnecessary information. The nature of such information depends on the use case and the AI / ML algorithm.

[0019] The model inference function should only signal model outputs to nodes that explicitly request them (e.g., via subscription) or nodes that take action based on the model inference outputs.

[0020] The AI / ML models used in the model inference function must be initially trained, validated, and tested before deployment.

[0021] RAN intelligent functional framework such as Figure 3 As shown. Data collection 302 is a function that provides input data for model training 304 and model inference 306. No AI / ML algorithm-specific data preparation (e.g., data preprocessing and cleaning, formatting, and transformation) is performed in the data collection function. Examples of input data may include measurements from the UE or different network entities, feedback from the actor, and outputs from the AI / ML model. Training data includes the data required as input to the AI / ML model training function. Inference data includes the data required as input to the AI / ML model inference function.

[0022] Model Training (304) is the function that performs ML model training, validation, and testing. It can generate model performance metrics as part of the model testing process. If needed, the Model Training function is also responsible for data preparation (e.g., data preprocessing and cleaning, formatting, and transformation) based on the training data delivered by the Data Collection function. Model Deployment / Update is used to initially deploy trained, validated, and tested AI / ML models to the Model Inference function, or to deliver updated models to the Model Inference function.

[0023] Model Inference 306 is a function that provides AI / ML model inference outputs (such as predictions or decisions). The Model Inference function can provide model performance feedback to the model training function where applicable. If necessary, the Model Inference function is also responsible for data preparation (e.g., data preprocessing and cleaning, formatting, and transformation) based on the inference data delivered by the data collection function. The output includes the inference output of the AI / ML model generated by the Model Inference function. Details of the inference output are use-use specific. Model performance feedback can be used to monitor the performance of the AI / ML model.

[0024] Actor 308 is a function that receives the output of the model's inference function and triggers or executes corresponding operations. The executor can trigger operations against other entities or itself. Feedback includes information needed to derive training or inference data or performance feedback.

[0025] RAN Intelligence has identified three use cases: network power saving, load balancing, and mobility optimization, the potential standard impact of which is described below in the same TR 37.817.

[0026] For network power saving use cases: AI / ML model training resides in the Operations, Administration, and Maintenance (OAM) function, while AI / ML model inference resides in the gNB; or both AI / ML model training and AI / ML model inference reside in the gNB. Note: It is also permissible for the gNB to continue model training based on the AI / ML model trained in the OAM. In the case of a CU-DU separation architecture, the following solutions are possible: AI / ML model training resides in the OAM, while AI / ML model inference resides in the gNB-CU; or both AI / ML model training and model inference reside in the gNB-CU.

[0027] For mobility optimization use cases, the location of AI / ML model training and inference can be considered in two ways: the AI / ML model training function is deployed in OAM, and the model inference function is located within the RAN node; or both the AI / ML model training and inference functions are located within the RAN node. Furthermore, for CU-DU separation scenarios, the following options are possible: AI / ML model training is located in CU-CP or OAM, and the AI / ML model inference function is located in CU-CP. Note: gNB is also allowed to continue model training based on the AI / ML model trained in OAM.

[0028] For load balancing use cases, the following solutions can be considered to support AI / ML-based load balancing: AI / ML model training resides in the OAM, and AI / ML model inference resides in the gNB; or both AI / ML model training and AI / ML model inference reside in the gNB. In a CU-DU separation architecture, the following solutions are possible: AI / ML model training resides in the OAM, and AI / ML model inference resides in the gNB-CU, or both AI / ML model training and model inference reside in the gNB-CU. Note: It is also permissible for the gNB to continue model training based on the AI / ML model trained in the OAM.

[0029] At the RAN3#114-e meeting, proposal R3-215244 proposed introducing model management functionality into the RAN Intelligent Functions Framework, as described below.

[0030] Figure 4 The functional framework of RAN intelligence is illustrated. For example... Figure 4As shown, model deployment / updates can be decided by model management 410 instead of model training 404. Model management 410 can also host model repositories. Model deployment / updates should be performed by model management. Model performance monitoring is a key function to assist and control model inference 406. Model performance feedback from model inference 406 should first be sent to model management 410. If performance is unsatisfactory, model management 410 can decide to fall back to a traditional algorithm or change / update the model.

[0031] In this scenario, model training should also be controlled by model management. Model management functionality can be handled by OAM, CU, or other network entities depending on the use case. Clearly defining model management functionality will facilitate future signaling design and analysis.

[0032] The model management function supports the following roles: requesting model training and receiving model training results; deploying / updating models for inference; monitoring model performance, including receiving performance feedback on model inference and taking necessary actions, such as retaining the model, rolling back to the traditional algorithm, changing or updating the model; and model storage. Summary of the Invention

[0033] Designing general-purpose AI applications for communication networks requires considering the various intents driving network functions. In RAN systems, performance is not defined by a single Key Performance Indicator (KPI), but rather by multiple quantifiable and often competing KPIs. For example, 3GPP TS 28.312 (such as v. 18.0.0) describes the concept of intent as a tool for specifying the goals, requirements, and constraints of service or network management workflows. Therefore, RAN control problems such as Radio Resource Management (RRM), including user and resource scheduling, do not have a single RAN-optimal KPI, but rather a Pareto front of RAN KPIs, with trade-offs between them. One example is the trade-off between optimizing spectral efficiency or throughput and transmission reliability or latency.

[0034] Most AI research in communication networks has proposed algorithms for optimizing specific intent functions at runtime, particularly for radio resource allocation and link adaptation for user equipment. Handling different intents at runtime requires training multiple models for each RAN function (e.g., user scheduling or link adaptation), which presents significant challenges. First, developing a unique AI model for each RAN function (e.g., link adaptation) to optimize specific intents (such as specific reward functions in reinforcement learning methods) requires training and managing multiple AI models to handle and optimize the performance of different intents (e.g., reward functions) at runtime. This is expensive and impractical even for fully reliable and automated systems. Second, dedicated AI models for optimizing specific intents reduce the system's robustness in handling and adapting to changes that may render the specific intent infeasible. This can lead to the system operating in a suboptimal configuration, potentially resulting in resource over-provisioning and performance degradation.

[0035] This disclosure provides a method for a first network node to optimize transmission parameters in the downlink or uplink of at least one user device based on at least one intent associated with multiple user devices. This disclosure also provides a method for training and executing a single AI / ML model that, at runtime, can select transmission parameters for user devices to optimize multiple intents, thereby improving system performance and scalability. At runtime, the intents used to select transmission parameters can be user-specific or user group-specific. The intents associated with a user device or a group of user devices can be represented, for example, by an individual-related objective function, utility function, performance metric, key performance indicators, and quality of service and quality of experience requirements. When multiple intents are associated with a user device or a group of user devices, this method provides a solution to find the optimal or near-optimal tradeoffs for achieving the intents.

[0036] One advantage of the aspects of this disclosure is that it enables the first network node to dynamically optimize and adjust multiple transmission parameter configurations of the user equipment, such as radio resource allocation and link adaptation parameters, for one or more communication intentions or goals associated with the user equipment. Therefore, system resources can be optimized to meet the needs of individual user equipment, thereby improving system spectral efficiency and system performance.

[0037] Another advantage of the aspects of this disclosure is that it enables the first network node to optimize multiple (often competing) communication intentions for each individual user device using a single solution (such as a single AI / ML model). This reduces the complexity of the first network node compared to situations requiring multiple separate solutions to optimize the different communication intentions of user devices. Therefore, the first network node can more efficiently determine the configuration of multiple transmission parameters for user devices, thereby achieving, for example, energy savings.

[0038] According to one aspect, a computer-implemented method, executed by a first network node in a communication network environment, is provided for optimizing transmission parameters for at least one user equipment. The method includes acquiring a first set of at least one intent of the at least one user equipment. The method includes acquiring at least one intent preference value associated with the at least one user equipment based on the acquired first set of at least one intent. The method includes determining a communication configuration of the at least one user equipment based on the at least one intent preference value using a first AI / ML model. The method includes initiating a change in the configuration of the at least one user equipment based on the communication configuration.

[0039] In some embodiments, the first set of at least one intent includes one or more of the following: objective function, utility function, performance metric, key performance indicator (KPI), quality of service (QoS) requirement, or quality of experience (QoE) requirement.

[0040] In some embodiments, the objective function, utility function, performance metric, and KPI include one or more of the following: user throughput, user spectral efficiency, information transmitted per packet, resource utilization or allocation, delay, block error rate, bit error rate, reliability, packet error rate, energy consumption, energy saving, discontinuous transmission, discontinuous reception, or signal quality, and wherein the QoS or QoE requirement includes one or more of the following: throughput requirement, spectral efficiency requirement, transmission delay requirement, block error rate requirement, packet error rate requirement, bit error rate requirement, packet error rate requirement, reliability requirement, signal-to-noise ratio requirement, resource utilization requirement, energy consumption requirement, energy saving requirement, resource utilization requirement, discontinuous transmission requirement, discontinuous reception requirement, transport block size requirement, or resource element requirement.

[0041] In some embodiments, the communication configuration includes one or more of the following: allocation or configuration of at least one communication resource, allocation or configuration of link adaptation parameters, allocation or configuration of information bits to be received or transmitted by the at least one user equipment, allocation or configuration of downlink or uplink reference signals, configuration for discontinuous transmission or reception, or configuration for energy saving.

[0042] In some embodiments, a first set of at least one intent is contained in a first message sent by the at least one user device or a second network node.

[0043] In some embodiments, the method further includes sending a second message to the at least one user equipment or the second network node, the second message including a request for a first set of intents.

[0044] In some embodiments, the method further includes sending a third message to the at least one user equipment, the third message including communication configuration.

[0045] In some embodiments, obtaining the at least one intent preference value includes mapping a first set of the at least one intent to the at least one intent preference value. In some embodiments, the mapping includes mapping the first set of the at least one intent to one or more user performance metrics parameterized relative to the one or more intent preference values. In some embodiments, the mapping is performed by obtaining user performance metrics parameterized with respect to the one or more intent preference values ​​by testing a first AI / ML model within a range of values ​​assumed for the one or more intent preference values. In some embodiments, the mapping is performed by a second AI / ML model.

[0046] In some embodiments, the first set of at least one intent includes a first intent and a second intent, and wherein at least one intent preference value includes a first intent preference value associated with the first intent and a second intent preference value associated with the second intent. In some embodiments, the first intent and the second intent correspond to a first user equipment. In some embodiments, the first intent corresponds to a first user equipment, and the second intent corresponds to a second user equipment different from the first user equipment. In some embodiments, the first set of at least one intent includes at least one intent related to transmission reliability, throughput, spectral efficiency, block error rate, or latency. In some embodiments, the first intent corresponds to the number of information bits carried in a data packet transmission to or from the at least one user equipment, the second intent corresponds to the total number of resource elements used for communication with the at least one user equipment up to the nth transmission attempt of the data packet, and the communication configuration includes one or more link adaptive parameters for downlink or uplink transmission.

[0047] In some embodiments, each component of the at least one intent r j Contained in an N-dimensional reward function vector r In, and wherein each component of the at least one intention preference value w j With the reward function vector r corresponding components r jAssociated. In some embodiments, determining using the first AI / ML model includes providing the first AI / ML model with at least one of the following: a set of one or more information elements indicating or characterizing the state of the first network node, the state of the at least one user device, the state of the communication network environment related to communication between the first network node and the at least one user device, or a combination thereof, an indication of the at least one intent, an indication of at least one intent preference value to be optimized for the at least one user device, or a combination thereof.

[0048] In some embodiments, the method further includes acquiring at least one training data sample associated with a first AI / ML model or a second AI / ML model. In some embodiments, the method further includes training the first AI / ML model based on the at least one training data sample; or sending a fourth message to a third network node including the at least one training data sample. In some embodiments, the method further includes receiving a fifth message, the fifth message including the first AI / ML model.

[0049] According to another aspect, a network node in a communication network environment is provided for optimizing the transmission parameters of at least one user equipment, the node being configured to perform the above-described method.

[0050] According to another aspect, a node in a communication network environment is provided for optimizing transmission parameters of at least one user equipment. The node includes: one or more memories, including instruction data representing an instruction set; and one or more processors configured to communicate with the one or more memories and execute the instruction set, wherein the instruction set, when executed by the processors, causes the one or more processors to perform the methods described above.

[0051] According to another aspect, a computer-implemented method, executed by a first user equipment in a communication network environment, is provided for optimizing transmission parameters of the user equipment. The method includes generating a first message, the first message including a first set of at least one intent of the user equipment. The method includes sending the first message to a first network node. The method includes obtaining a communication configuration. The method includes updating one or more transmission parameters based on the communication configuration.

[0052] In some embodiments, the method further includes receiving a second message, the second message including a request for a first set of the at least one intent of the user equipment.

[0053] In some embodiments, the first set of at least one intent includes one or more of the following: objective function, utility function, performance metric, key performance indicator (KPI), quality of service (QoS) requirement, or quality of experience (QoE) requirement.

[0054] In some embodiments, the objective function, utility function, performance metric, and KPI include one or more of the following: user throughput, user spectral efficiency, information transmitted per packet, resource utilization or allocation, delay, block error rate, bit error rate, reliability, packet error rate, energy consumption, energy saving, discontinuous transmission, discontinuous reception, or signal quality, and wherein the QoS or QoE requirement includes one or more of the following: throughput requirement, spectral efficiency requirement, transmission delay requirement, block error rate requirement, packet error rate requirement, bit error rate requirement, packet error rate requirement, reliability requirement, signal-to-noise ratio requirement, resource utilization requirement, energy consumption requirement, energy saving requirement, resource utilization requirement, discontinuous transmission requirement, discontinuous reception requirement, transport block size requirement, or resource element requirement.

[0055] In some embodiments, the communication configuration includes one or more of the following: allocation or configuration of at least one communication resource, allocation or configuration of link adaptation parameters, allocation or configuration of information bits to be received or transmitted by the at least one user equipment, allocation or configuration of downlink or uplink reference signals, configuration for discontinuous transmission or reception, or configuration for energy saving.

[0056] According to another aspect, a user equipment in a communication network environment is provided for optimizing transmission parameters of a user equipment. The user equipment includes one or more memories, including instruction data representing a set of instructions; and one or more processors configured to communicate with the one or more memories and execute the set of instructions, wherein, when executed by the processors, the set of instructions causes the one or more processors to perform the method 20 described above.

[0057] In another aspect, a user equipment in a communication network environment is provided for optimizing the transmission parameters of the user equipment, the user equipment being configured to perform the above-described method.

[0058] In another aspect, a computer program containing instructions is provided, which, when executed by processing circuitry, cause the processing circuitry to perform the described method. In yet another aspect, a carrier containing a computer program is provided, wherein the carrier includes one of electronic signals, optical signals, radio signals, or a computer-readable storage medium. In yet another aspect, a computer program product is provided, comprising a non-transitory computer-readable medium on which the computer program is stored. In yet another aspect, an apparatus is provided, comprising: a memory; and processing circuitry coupled to the memory, wherein the apparatus is configured to perform the described method. Attached Figure Description

[0059] The accompanying drawings, which are incorporated herein and form part of the specification, illustrate various embodiments.

[0060] Figure 1 The NG-RAN node architecture is shown.

[0061] Figure 2 The NG-RAN node architecture is shown.

[0062] Figure 3 The functional framework of RAN intelligence is shown.

[0063] Figure 4 The functional framework of RAN intelligence is shown.

[0064] Figure 5 Methods according to some embodiments are shown.

[0065] Figure 6 Signaling diagrams according to some embodiments are shown.

[0066] Figure 7 Signaling diagrams according to some embodiments are shown.

[0067] Figure 8 This is a flowchart based on some embodiments.

[0068] Figure 9 This is a flowchart based on some embodiments.

[0069] Figure 10 This is a flowchart based on some embodiments.

[0070] Figure 11 A flowchart according to some embodiments is shown.

[0071] Figure 12 A graph is shown according to some embodiments.

[0072] Figure 13 A graph is shown according to some embodiments.

[0073] Figure 14 A graph is shown according to some embodiments.

[0074] Figure 15 A flowchart according to some embodiments is shown.

[0075] Figure 16 Methods according to some embodiments are shown.

[0076] Figure 17 Methods according to some embodiments are shown.

[0077] Figure 18 It is a block diagram of a network node or user equipment according to some embodiments. Detailed Implementation

[0078] As used in this document, network nodes can be RAN nodes, OAM, core network nodes, OAM, Service Management and Orchestration Function (SMO), Network Management System (NMS), Non-RT RAN Intelligent Controller (Non-RT RIO), Real-time RAN Intelligent Controller (RT-RIC), gNB, eNB, en-gNB, ng-eNB, gNB-CU, gNB-CU-CP, gNB-CU-UP, eNB-CU, eNB-CU-CP, eNB-CU-UP, IAB nodes, IAB donor DU, IAB donor CU, IAB-DU, IAB-MT, O-CU, O-CU-CP, O-CU-UP, O-DU, O-RU, O-eNB, and UE.

[0079] Unless otherwise explicitly stated, the terms model training, model optimization, model optimization, and model update are used interchangeably in this article and have the same meaning.

[0080] Unless otherwise expressly stated, the terms "model change," "modification," or similar expressions are used interchangeably throughout this document and have the same meaning. In particular, these phrases refer to the fact that the type, structure, parameters, connectivity, or other aspects of the AI / ML model may have changed compared to the previous format / configuration of the AI / ML model.

[0081] Unless otherwise expressly stated, the terms AI / ML model, AI / ML strategy, AI / ML algorithm, and the terms model, strategy or algorithm are used interchangeably in this document and have the same meaning.

[0082] The term "network node" as used in this article should be understood as any type of physical node, functional or logical entity, such as a software entity implemented in a data center or cloud, for example, using one or more virtual machines, and two network nodes can also be implemented as logical software entities in the same data center or cloud.

[0083] Non-limiting examples of AI / ML algorithms can include supervised learning algorithms, deep learning algorithms, reinforcement learning-type algorithms (such as DQN, A2C, A3C, etc.), context-based multi-armed slot machine algorithms, autoregressive algorithms, etc., or combinations thereof. These algorithms can utilize function approximation models (hereinafter referred to as AI / ML models), such as neural networks (e.g., feedforward neural networks, deep neural networks, recurrent neural networks, convolutional neural networks, etc.). Examples of reinforcement learning algorithms can include deep reinforcement learning (such as Deep Q-Networks (DQN), Proximal Policy Optimization (PPO), Double Q Learning), actor-commentator algorithms (such as dominant actor-commentator algorithms, such as A2C or A3C, actor-commentator algorithms with experience replay, etc.), policy gradient algorithms, off-policy learning algorithms, etc.

[0084] Unless otherwise expressly stated, the terms “intent,” “intent function,” “reward,” and “reward function” are used interchangeably herein and have the same meaning. An intent value or reward value indicates a quantifiable value assumed by an intent or reward. The term “intent vector” or “reward vector” refers to a set of one or more intents, intent functions, rewards, or reward functions, each indicating a quantifiable value. In some embodiments, an intent may be defined by multiple quantifiable and often competing KPIs. For example, 3GPP TS 28.312 (such as v.18.0.0) describes the concept of intent as a tool for specifying the goals, requirements, and constraints of service or network management workflows.

[0085] Figure 5 illustrates a method according to some embodiments. Method 500 can be performed by a first network node to optimize transmission parameters of at least a first user equipment in a radio communication system based on multiple intentions associated with a user equipment and / or at least a second user equipment. Step s501 of the method includes obtaining a set of at least one intention for optimizing communication to or from at least one user equipment. Step s503 of the method includes obtaining at least one intention preference value associated with at least one user equipment based on the corresponding at least one intention for optimizing communication to or from the user equipment. Step s505 of the method includes determining a communication configuration of the user equipment based on the at least one intention preference value associated with the user equipment. Step s507 of the method includes configuring at least one user equipment to send information to or receive information from the first network node based on optimizing the communication configuration associated with at least one intention.

[0086] In some embodiments, the intent associated with a user device may include any one or a combination of the following: an objective function, a utility function, a performance metric, a key performance indicator, or one or more of a service quality requirement or a experience quality requirement.

[0087] In some embodiments, the communication configuration of a user equipment (whether downlink or uplink) may include multiple transmission parameters, such as, for example, one or a combination of the following: allocation of at least one communication resource, allocation of link adaptation parameters, allocation of information bits, or allocation or configuration of downlink or uplink reference signals.

[0088] In some embodiments, the term "intent" refers to a quantifiable value based on one or more of the following: objective function, utility function, performance metric, key performance indicator, and quality of service requirements, quality of experience requirements, traffic requirements, etc. Communication configuration can represent a first network node determining one or more transmission parameters to optimize user equipment communication based on associated communication intents and / or intent preferences.

[0089] In some embodiments, step s505 of determining the communication configuration of the user equipment based on at least one intention preference value associated with the user equipment may include determining one or more of the following: (i) the configuration or allocation of at least one communication resource (e.g., for downlink or uplink transmission), such as at least one physical resource block (PRB), at least one physical resource block group (PRBG), at least one sub-frequency band, at least one transmission time interval; (ii) the configuration or allocation of at least one link adaptation parameter (e.g., for downlink or uplink transmission), such as communication and coding scheme, modulation order, transmission rank; (iii) the configuration or allocation of information bits to be received or transmitted by the user equipment (i.e., for downlink or uplink transmission, respectively); (iv) the configuration or allocation of downlink or uplink reference signals for communication with the user equipment; (v) the configuration for discontinuous transmission or discontinuous reception to optimize, for example, power saving of the user equipment; and / or (vi) power saving configurations, such as configurations for discontinuous transmission and / or discontinuous reception, configurations for activating or deactivating sleep modes, and instructions for entering the RRC_DLE state.

[0090] Figure 6 A signaling diagram according to some embodiments is shown. In some embodiments, step s501 of obtaining a set of at least one intent for optimizing communication with at least one user equipment and / or step s503 of obtaining at least one intent preference value associated with at least one user equipment may require a first network node 600 to receive a first message 603 from at least one user equipment 602, the first message indicating information related to at least one intent for optimizing communication (i.e., sending and / or receiving) with at least one user equipment.

[0091] like Figure 6 As shown, when the first network node 600 receives information related to the user equipment's intent directly from the user equipment 602, the first network node may additionally send a second message 605 to the user equipment. The second message requests information related to at least one intent for optimizing communication (i.e., sending and / or receiving) with at least one user equipment.

[0092] At point 607, the first network node 600 performs intent-based user communication optimization, as described in further detail below.

[0093] In some embodiments, configuring at least one user equipment to send information to or receive information from a first network node based on an optimized communication configuration for at least one intent associated with the user equipment may include sending a third message 609 to the user equipment, the third message providing the user equipment with an optimized communication configuration for one or more intents associated with the user equipment.

[0094] In some embodiments, the first message 603 may be transmitted by the user equipment as part of an existing or new Radio Resource Control (RRC) type message. In one example, the first network node may receive the first message or one or more pieces of information provided by the first message as part of a UE capability transmission procedure defined by the 3GPP NG-RAN or E-UTRAN system. In one example, the first network node may receive the first message or one or more pieces of information provided by the first message as part of a UECapabilityInformation message. Therefore, information related to at least one intent to optimize communication (i.e., transmission and / or reception) with at least one user equipment may be included in the UECapabilityInformation message. Additionally, the first network node may send a second message 605 or one or more pieces of information provided by the second message as part of a UECapability Enquiry message defined by the 3GPP NG-RAN or E-UTRAN system.

[0095] Figure 7 Signaling diagrams according to some embodiments are shown. In such... Figure 7 In another signaling variant shown, obtaining a set of at least one intent for optimizing communication with at least one user equipment and / or obtaining at least one intent preference value associated with at least one user equipment may require the first network node 700 to receive a first message 703 from the second network node 704, the first message indicating information related to at least one intent for optimizing communication (i.e., sending and / or receiving) with at least one user equipment 702.

[0096] In some embodiments, the second network node 704 may represent the source node, while the first network node 700 may represent the target node involved in the handover process of user equipment 702. In other words, user equipment 702 may move between radio cells controlled by the second network node 704 and radio cells controlled by the first network node 700. In one possible implementation of this embodiment, the radio resource control (RRC) signaling anchor point of the user equipment may be moved from the second network node to the first network node. In another possible implementation of this embodiment, the radio resource control (RRC) signaling anchor point of the user equipment may be retained by the second network node. In this case, the first network node may be the control node serving a new secondary cell for the user equipment, such as in the case of a dual-connectivity process.

[0097] Using this signaling variant, the first network node 700 can receive information from the second network node 704 related to at least one intent for optimizing communication (i.e., sending and / or receiving) with at least one user equipment 702. In some cases, such communication may involve serving the user equipment through joint sending or receiving capabilities with a second or third network node. Therefore, the management of user equipment intents may need to be coordinated or shared among multiple network nodes. Coordinating the management of user equipment intents among multiple network nodes can provide advantages, such as reducing the risk of user equipment being poorly configured relative to multiple network nodes. A potential advantage of this signaling variant is that it can ensure efficient and apparent fulfillment of user intents (such as quality of service requirements) during mobility events between two network nodes. An additional potential advantage of this signaling variant is that it can ensure that multiple transmission points providing joint sending or joint receiving type communication for the user equipment can efficiently fulfill user intents.

[0098] In the event of a handover event, the first network node 700 may receive a first message 703 from the second network node as part of a handover request message. Therefore, the second network node may provide information related to at least one intent for optimizing communication (i.e., sending and / or receiving) with at least one user equipment as part of the handover request message. In an alternative embodiment, the first network node may receive the first message 703 as part of a user equipment context setting procedure. In this case, the first message 703 may be provided by the second network node to the first network node as part of an initial UE context setting request message for, for example, a 3GPPNG-RAN or E-UTRAN system. In this case, the first network node represents a RAN node, while the second network node may represent a managed access and mobility management function (AMF) node.

[0099] exist Figure 7 In the signaling variant shown, the first network node receives information related to the user equipment's intent from the second network node. The first network node 700 can send a second message 705 to the second network node 704, which requests information related to at least one intent for optimizing communication (i.e., sending and / or receiving) with at least one user equipment. In some embodiments, the first network node can receive a fourth message from the second network node (…). Figure 7 (not shown in the image), the second message indicates a positive acknowledgment (ACK) or a negative acknowledgment (NACK) related to information relating to the intent requested by the first network node.

[0100] At point 707, the first network node 700 performs intent-based user communication optimization, as described in further detail below.

[0101] In some embodiments, configuring at least one user equipment to send information to or receive information from a first network node based on an optimized communication configuration for at least one intent associated with the user equipment may include: the first network node 700 sending a third message 709 to a second network node 704 or the user equipment 702, the third message providing the user equipment with an optimized communication configuration for one or more intents associated with the user equipment. In one example, the second network node may additionally forward the third message or information received in the third message to the user equipment.

[0102] In some embodiments, a first network node may receive a first message 603, 703 from a second network node and / or at least one user equipment that needs to optimize sending or receiving parameters to satisfy one or more intents. In one embodiment, the information provided in the first message is related to at least one intent for optimizing communication (i.e., sending and / or receiving) with at least one user equipment. The first message may include one or more of the following: at least one intent for optimizing communication with the user equipment and / or at least one intent preference value associated with at least one intent indicated for optimizing communication with the user equipment.

[0103] In some embodiments, the intent indicated by the first message associated with the user equipment may include one or more objective functions, utility functions, performance metrics and key performance indicators and / or a combination of one or more quality of service requirements and quality of experience requirements.

[0104] Non-restricted examples of the objective function, utility function, performance metric, and key performance indicator indicated as the intent in the first message may include one or more of the following: User throughput. Non-limiting examples may include: (a) an indication to maximize user throughput; (b) an indication to maximize peak user throughput; (c) an indication to maximize average user throughput; (d) an indication to maximize minimum user throughput; (e) an indication of a range of user throughput to be provided; and / or (f) an indication of a minimum user throughput to be provided.

[0105] User spectral efficiency. Non-limiting examples may include: (a) an indication of maximizing spectral efficiency; (b) an indication of maximizing peak spectral efficiency; (c) an indication of maximizing average spectral efficiency; (d) an indication of maximizing minimum spectral efficiency; (e) an indication of a range of user spectral efficiency provided; and (f) an indication of the minimum user spectral efficiency to be provided.

[0106] Information transmitted in each data packet. Non-limiting examples may include indications of transport block size, MAC-PDU size, PDCP-PDU size, etc., as defined by the 3GPP LTE or NG-RAN system.

[0107] Resource utilization or allocation. In a non-limiting example, resource utilization or allocation may be expressed as a function of one or more of the following: (a) the number of resource elements (REs) transmitted per packet, such as resource elements of an OFDM-based time-frequency resource grid in a 3GPP E-UTRAN or NG-RAN system, consisting of subcarriers in the frequency domain and OFDM symbols in the time domain; (b) the number of aggregated resource elements (REs) allocated or used per packet transmission, wherein the aggregation includes the number of resource elements used in multiple transmissions of the same packet; (c) the number of physical resource blocks (PRBs) or physical resource block groups (PRBGs) allocated or used per packet transmission, such as those defined by an OFDM-based time-frequency resource grid in a 3GPP E-UTRAN or NG-RAN system; and / or (d) the number of aggregated resource elements transmitted per packet, wherein the aggregation includes the number of resource elements used in multiple transmissions of the same packet.

[0108] Delay. In a non-restrictive example, delay can be expressed as a function of one or more of the following: (a) the delay of each packet transmission; (b) the delay of each packet successful delivery; (c) the end-to-end delay across layers; and / or (d) the data service delay, etc.

[0109] Block Error Rate (BLER). In a non-restrictive example, BLER can be expressed as a function of one or more of the following: (a) BLER per packet transmission, and / or (b) statistical BLER, such as the mean, variance, or standard deviation of multiple packets.

[0110] Bit Error Rate (BER). In a non-restrictive example, BER can be expressed as a function of one or more of the following: (a) the BER per packet transmitted; and / or (b) a statistical BER, such as the average, variance, or standard deviation of multiple packets.

[0111] Reliability. In a non-restrictive example, reliability can be expressed as a function of one or more of the following: (a) reliability of each packet transmission; (b) statistical reliability, such as the average, variance, or standard deviation of multiple packets; (c) end-to-end reliability across layers; (d) data service reliability, etc.

[0112] Additional examples may include packet error rate, energy consumption, energy saving, discontinuous transmission, discontinuous reception and / or signal quality, such as signal-to-interference-plus-noise ratio or signal-to-noise ratio.

[0113] Non-limiting examples of service quality requirements and experience quality requirements indicated as intent in the first message may include one or more of the following: one or more throughput requirements, such as minimum throughput requirements; one or more spectral efficiency requirements, such as minimum spectral efficiency requirements; one or more transmission delay requirements, such as maximum transmission delay requirements; one or more block error rate requirements, such as maximum block error rate requirements; one or more packet error rate requirements, such as maximum packet error requirements; one or more bit error rate requirements, such as maximum bit error rate requirements; one or more packet error rate requirements, such as maximum packet error requirements; one or more reliability requirements, such as minimum reliability requirements; one or more signal-to-interference-plus-noise ratio (SIR) requirements, such as minimum SIR; one or more SNR requirements, such as minimum SNR requirements; one or more resource utilization requirements, such as minimum resource utilization requirements; one or more energy consumption requirements, such as maximum energy consumption requirements; one or more energy saving requirements, such as minimum energy saving requirements; one or more resource utilization requirements, such as minimum resource utilization requirements; one or more discontinuous transmission or discontinuous reception requirements, such as minimum or maximum discontinuous transmission or discontinuous reception requirements; one or more transport block size requirements, such as minimum or maximum transport block size requirements; and / or one or more resource element number requirements, such as minimum or maximum resource element number requirements.

[0114] In some embodiments, third messages 609, 709 provide communication configuration for the user equipment, optimizing one or more intentions associated with the user equipment. This communication configuration may include one or more of the following: configuration or allocation of at least one communication resource (e.g., for downlink or uplink transmission), such as at least one physical resource block (PRB), at least one physical resource block group (PRBG), at least one sub-band, at least one transmission time interval; configuration or allocation of at least one link adaptation parameter (e.g., for downlink or uplink transmission), such as communication and coding scheme, modulation order, transmission rank; configuration or allocation of information bits to be received or transmitted by the user equipment (i.e., for downlink or uplink transmission, respectively); configuration or allocation of downlink or uplink reference signals for communication with the user equipment; configuration for discontinuous transmission or discontinuous reception to optimize, for example, power saving by the user equipment; and / or power saving configurations, such as configurations for discontinuous transmission and / or discontinuous reception, configurations for activating or deactivating sleep modes, and instructions to enter the RRC_DLE state.

[0115] In some embodiments, obtaining at least one intent preference value associated with at least one user device may include one or more of the following: (i) receiving a first message from the user device or from a second network node, the first message including at least one intent for optimizing communication with the user device; and / or (ii) determining one or more intent preference values ​​associated with one or more intents based on the at least one intent for optimizing communication with the user device.

[0116] The first network node acquires and optimizes communications to or from user equipment. In embodiments involving a set of intents (N equal to or greater than 1), obtaining at least one intent preference value associated with a user device intent may include determining A set of N (N equal to or greater than 1) intent preference values, where each intent preference value represents the user device’s relative preference for the associated intent function.

[0117] Figure 8 This is a flowchart based on some embodiments. In, for example... Figure 8 In some embodiments shown, obtaining at least one intent preference value associated with a user device intent can be achieved through a function 800 (hereinafter referred to as the intent manager), which maps one or more user intents to one or more intent preference values. In the special case where the mapping is achieved through an identity function such as an identity matrix, the intent preference value corresponds to a user communication intent value. In a first embodiment, the intent manager 800, which maps user intents to intent preference values, is implemented by using parameterized user performance obtained from test results of a first AI / ML model used by a first network node to determine the communication configuration of at least one user device. In a second embodiment, the intent manager 800 is implemented by using a second AI / ML model trained to map user intents to intent preference values.

[0118] In some embodiments, the intent manager 800 can be implemented by mapping one or more user intents to one or more user performance parameters parameterized relative to one or more intent preference values. For example, user performance associated with different transport parameter configurations can be evaluated for multiple values ​​for each intent preference. In one example, user performance parameterized by different intent preference values ​​can be obtained by testing an AI / ML model (which is used interchangeably herein with "first AI / ML model") within a range of values ​​that can be assumed for the intent preference values. In some embodiments further described below, the AI / ML model is used to determine the communication configuration of a user device or a group of user devices. Examples of test results obtained by the AI / ML model will be further described below, which a first network node can use to map user intents to intent preference values.

[0119] In one example, with the intent function of the user device Associated intention preference value It can represent the user device relative to the intent function associated with the user device. For intention function Preferences.

[0120] For example, suppose the user device is equal to the intent function of throughput. The sum equals the intention function of communication reliability These two intent functions are associated with each other, and are respectively with and Different pairs of related preference values and This can lead to different choices in communication configuration.

[0121] if (For example, and In this case, the user equipment (UE) prefers reliable communication over high throughput. Therefore, under good channel conditions, the UE can be configured to communicate reliably even with high throughput parameters (e.g., with large modulation and coding scheme indices and / or high rank), while under poor channel conditions, the UE will be configured to communicate with reliable communication parameters, possibly at the expense of throughput (e.g., with low modulation and coding scheme indices and / or low rank).

[0122] if (For example, and If the channel conditions are such that the user equipment prefers high data rate communication over reliability, then the user equipment may always be configured to communicate with high throughput parameters, possibly at the expense of reliable communication (e.g., with large modulation and coding scheme indices and / or high rank).

[0123] In an unrestricted implementation, the set of N intent preference values ​​can be represented as an N-dimensional preference vector associated with the user device, based on the user device intent. In one example, the preference vector Each component is a real value, that is... Each component This indicates the user device's relative preference for the corresponding intent function associated with it.

[0124] In some embodiments, obtaining at least one intent preference value associated with at least one user device may include one or more of the following: determining one or more intent preference values ​​associated with one or more intents of the user device based on (a) one or more intents of the user device and / or (b) one or more intents of at least a second user device. In this case, the user device’s preference for a particular intent is evaluated relative to the preference of at least the second user device for the same intent or other intents.

[0125] In a non-restrictive example, a network node may acquire or determine: (i) with the first intent function (For example, throughput) associated with the first user device's first intent preference value , and (ii) with the second intention function (For example, reliability) The second intention preference value of the second user device. .

[0126] In one implementation of this example, the intent preference value and This can represent the relative preference or importance that the first network node should consider when allocating radio resources to the first user equipment and the second user equipment. In one example, where the first network node can allocate resources to both the first and second user equipment simultaneously, the first network node can: (i) if Then, radio resources are allocated preferentially to the first user equipment, or (ii) if In this case, radio resources will be allocated to the second user equipment first.

[0127] In some embodiments, the first network node determines the communication configuration of the user equipment based on a multi-objective reinforcement learning algorithm, which is based on at least intent preference values ​​associated with the user equipment and an AI / ML model. In some embodiments, the first network node executes a K-dimensional reward function vector-based... A reinforcement learning algorithm, where K is equal to or greater than 1, and one or more reward components. It can correspond to one or more intentions. Such as objective functions, utility functions, performance metrics and key performance indicators and / or one or more service quality requirements or experience quality requirements indicated as intent in the first message.

[0128] In some embodiments, each reward component Corresponding to an intent indicated by the first message This increases the dimension of the reward function vector. It equals the dimension of the set of N intentions indicated by the first message, i.e. equal For example, the reward component can be the same as the intent (such as throughput), and there is a one-to-one mapping between the reward component and the intent.

[0129] In some embodiments, at least components With at least two intentions indicated by the first message This correlation, thus affecting the dimension of the reward function vector. The dimension is smaller than the set of N intentions indicated by the first message, that is, Less than For example, the reward component can implicitly represent an intent, such as using transport block size (TBS) instead of throughput.

[0130] In some embodiments, negative values ​​of one or more reward components are used instead of corresponding positive values. This allows for minimizing rather than maximizing the intent associated with reward components used with negative values.

[0131] In some embodiments, the first network node executes a reinforcement learning algorithm comprising an AI / ML model and a multidimensional reward function vector associated with packet transmissions to or from a user equipment. This multidimensional reward function vector includes: (i) an indication of the number of information bits carried by the packet, such as the transport block size (TBS) defined by the 3GPP E-UTRAN or NG-RAN system; and (ii) negative values ​​indicating the potential aggregate resource utilization used to deliver the packet in multiple transmission attempts. These negative values ​​may be one or more of the following: (a) the total number of resource elements. (a) Negative values, (b) Total number of physical resource blocks The negative value of (c) the total number of physical resource block groups. Negative values, or (d) total number of resource subbands The negative value.

[0132] In one example, the multidimensional reward function vector associated with packet transmission to or from a user device can be represented as shown in equation (1): Equation (1)

[0133] in, A collection of one or more information elements indicating or characterizing the state of a first network node and / or the state of a user equipment and / or the state of the radio environment related to communication between the first network node and the user equipment. Non-limiting examples of this information may include: (a) information characterizing the radio links in the uplink and / or downlink between the network node and the user equipment; (b) information characterizing radio interference that the user equipment may experience in the uplink or downlink; (c) information characterizing the resource utilization, load, or capacity of the user equipment in at least the serving cell; (d) information characterizing the resource utilization, load, or capacity of the user equipment in at least the interfering cell; (e) the type of traffic requested by the user equipment; and / or (f) the location of the user equipment relative to the serving cell and / or at least the interfering cell.

[0134] : Represents one or more communication configuration parameters, such as transmit or receive parameters, determined by the first network node based on one or more intentions for communicating with the user equipment. The communication configuration parameters may include one or more of the following: (a) allocation of at least one communication resource for downlink or uplink transmission, (b) allocation of link adaptation parameters for downlink or uplink transmission, (c) allocation of information bits for downlink or uplink transmission, and (d) allocation or configuration of downlink or uplink reference signals for communicating with the user equipment.

[0135] TBS represents the transport block size or the number of transport bits carried by data packets transmitted to user equipment.

[0136] This represents the data packets used to communicate with the user equipment up to the data packets associated with the reward function. The total number of resource elements up to the point of transmission attempt.

[0137] Therefore, the first network node can use a 2D reward function with the following values: (i) if in If a negative acknowledgment (NACK) is received at the packet transmission attempt, then a method with... (ii) if in the zero information bits of each resource element; and (ii) if in If a positive acknowledgment (ACK) is received at the packet transmission attempt, then the packet with the used... TBS information bits for each resource element.

[0138] In some embodiments, the first network node may additionally obtain an N-dimensional preference vector associated with the user equipment based on the user equipment's intent. , where the preference vector With The number of dimensions that have the same number of reward components, i.e. Each component This indicates the user device's response to the corresponding reward component. The relative preferences. The first network node can determine or parameterize the different reward components (i.e., objectives) associated with at least the user device with respect to a K-dimensional preference (importance) vector. Preference function of relative importance In each reward component Corresponding to an intent indicated by the first message In the embodiments, preference vector Each component It also indicates that the user equipment is responding to the corresponding intent. Relative preferences.

[0139] In the embodiment, the preference function This represents a linear combination of the weighted relative importance of the reward components, which is... .

[0140] In one embodiment of this method, the preference vector components must satisfy the condition that their cumulative sum equals one, i.e. .

[0141] Using a two-dimensional reward vector at the first network node (e.g., In one embodiment of the method, the first network node can obtain a preference vector associated with the user equipment, which is expressed as a single scalar parameter. The function takes values ​​within a continuous range between 0 and 1, i.e. In this case, the preference vector It can be represented as ,Right now Furthermore, the first network node can represent a preference function that is a linear combination of the weighted relative importance of reward components associated with the user device. Determined as .

[0142] In some embodiments, when the first network node uses a 2D reward function vector associated with packet transmissions of the user equipment (including a function of transport block size TBS and the utilization of negative cumulative resource elements for multiple transmissions of user packets), the first network node can determine a preference function representing a linear combination of the weighted relative importance of the reward components associated with the user equipment. for .

[0143] In such Figure 9 In some of the embodiments shown, the first network node determines the communication configuration for a user device or a group of user devices based on executing an AI / ML model 902. Figure 9 This is a flowchart based on some embodiments. The intent manager 900 can take one or more communication intents as input and output one or more intent preference values. The AI / ML model 902 can take the intent preference values ​​and possible user and / or network states as input and output user equipment communication configuration. In some embodiments, network states include the state of a first network node.

[0144] AI / ML model 902 may require one or more input information from the following group: (i) a set of one or more information elements indicating or characterizing the state of a first network node and / or the state of a user equipment and / or the state of the radio environment related to communication between the first network node and the user equipment; (ii) an indication of at least one communication intent to be optimized for at least one user equipment or group of user equipments; and / or (iii) an indication of at least one communication intent preference value to be optimized for at least one user equipment or group of user equipments. In one example, each communication intent preference value represents the user equipment's preference for an associated communication intent relative to other communication intents associated with the user equipment. In another example, each communication intent preference value represents the user equipment's preference for an associated communication intent relative to other communication intents associated with a second user equipment or a group of user equipments.

[0145] In some embodiments, the output of the AI / ML model 902 can directly provide communication configuration for a user equipment or a group of user equipments. In another embodiment, the output of the AI / ML model can indirectly provide communication configuration for a user equipment or a group of user equipments. In one example, the output of the AI / ML model can be used to determine one or more communication configuration parameters for a user equipment or a group of user equipments. Therefore, in some embodiments, the output of the AI / ML model can indicate the importance, correctness, or optimality of a particular communication configuration relative to other communication configurations for a user equipment.

[0146] The following describes a non-limiting example in which a first network node employs a multi-objective reinforcement learning algorithm using AI / ML Model 902 to optimize link adaptive transmission parameters based on one or more intents or intent preference values ​​associated with a user device. In this example, the 2D reward design is defined based on: (1) TBS: the number of information bits carried by the data packets transmitted to the user device (e.g., transport block size); and (2) Used to communicate with user equipment until the data packet associated with the reward function is received. The total number of resource elements attempted for transmission, the reward function is defined in the previous embodiment as follows: Equation (1)

[0147] Furthermore, in this example, the communication configuration parameters determined by the first network node based on at least one intent preference value associated with the user equipment include one or more link adaptive parameters for downlink or uplink transmission, in the following group: (i) for example, a modulation and coding scheme (MCS) index defined by the 3GPP technical specifications for E-UTRAN or NG-RAN systems, (ii) a modulation order, such as QPSK, 16QAM, 64QAM, 256QAM, 1024 QAM, etc., (iii) a transmission rank or number of spatial streams or spatial layers, and / or (iv) an allocation of information bits, such as a transport block size defined by the 3GPP technical specifications for E-UTRAN or NG-RAN systems.

[0148] The communication configuration parameters indicate that the first network node can determine one or more options to optimize the user equipment's communication intent preferences. In this example, the first network node utilizes intent-driven link adaptation to optimize communication with the user equipment.

[0149] In this example's unrestricted implementation, the first network node can be based on a single scalar parameter. Obtain or determine the preference value associated with each component of the reward vector, i.e. and Single scalar parameter The range of values ​​between 0 and 1, i.e. And respectively .

[0150] Therefore, the first network node can determine the preference parameters that provide the user equipment's relative preference for the first or second component of the reward function. To determine different link adaptive parameters for optimizing different intentions (such as transmission reliability, throughput, or spectral efficiency).

[0151] Figure 10 This is a flowchart based on some embodiments. Figure 10 A non-limiting implementation of the embodiment is shown, wherein the first network node executes AI / ML model 1002 as part of a multi-objective reinforcement learning algorithm, based on scalar... Parameterized intent preference values and To determine the minimum communication configuration parameters of the user equipment. Because the Intent Manager 1000 can... Figure 1 Harmony Figure 2 As input, and output respectively by scalar Parameterized intent preference values and This is to determine the user equipment communication configuration.

[0152] In some embodiments, the first network node may determine one or more communication configuration parameters for link adaptation, such as modulation order, coding rate, and transport block size, by: (1) mapping the desired user intent to intent preference values ​​using parameterized user performance obtained from test results of a first AI / ML model used by the first network node to determine the configuration communication parameters of the user equipment; and (2) determining the configuration communication parameters of the user equipment by performing an AI / ML model based at least on the user equipment and network state and the intent preference values ​​associated with the user equipment.

[0153] Figure 11 A flowchart according to some embodiments is shown. Figure 11 A non-limiting example is shown, in which the intent manager 1100 maps desired user intents to intent preference values ​​using parameterized user performance obtained from test results of an AI / ML model 1102 used by a first network node to determine user device configuration communication parameters. Figure 11 In a second, non-limiting example not shown, the intent manager 1100 maps desired user intents to intent preference values ​​using a second AI / ML model 1102.

[0154] Figure 12 and Figure 13 A graph is shown according to some embodiments. Figure 12-13 This illustrates different intent preference values ​​derived from the transmission parameters used to optimize downlink adaptation. A non-restricted example of parameterized user performance. In this example, Figure 12 It shows user throughput and spectrum efficiency, while Figure 13 The block error rate and latency (e.g., the average number of transmissions required per packet) obtained by testing the AI / ML model across multiple ranges of intent preference values ​​are shown to be applicable in the interval [0, 1]. . Figure 12-13 The example shown illustrates three operating points: ω=0 maximizes reliability and minimizes latency and resource utilization; ω=0.2 maximizes throughput, thus balancing latency and spectral efficiency; and ω=1 maximizes spectral efficiency.

[0155] In the first example, when the intent information associated with the user equipment needs to maximize transmission reliability or minimize packet transmission latency, the first network node can: (1) determine to use an intent preference value close to zero. (2) Determine one or more link adaptive parameters for optimizing reliability or latency for user equipment based on user equipment and network status and determined preference values.

[0156] By selecting preference values (That is, approximately zero), for this example, the first network node determines the link adaptive parameters that minimize the aggregate resource utilization used to deliver data packets to the user equipment.

[0157] In the second example, when the intent information associated with the user equipment needs to maximize transmission spectral efficiency (e.g., regardless of reliability and latency), the first network node can: (1) determine an intent preference value close to 1. (2) Determine one or more link adaptive parameters for optimizing reliability or latency for user equipment based on user equipment and network status and determined preference values.

[0158] By selecting preference values (That is, approximately one), the first network node determines the link adaptive parameters that maximize the transmission spectrum efficiency, although at the cost of a higher block error rate and latency (i.e., a larger number of retransmissions).

[0159] In the third example, when the intent information associated with the user device needs to maximize user throughput, the first network node can: (1) determine to use an intent preference value close to 0.2. (2) Determine one or more link adaptive parameters for optimizing reliability or latency for user equipment based on user equipment and network status and determined preference values.

[0160] Figure 12-13 This shows how to select preference values. (That is, approximately 0.2), the first network node determines the link adaptive parameters that maximize the throughput of the user equipment.

[0161] Figure 14 It is a graph based on some embodiments. Figure 14 The same example is illustrated by showing the Pareto front for two reward components (i.e., the transport block size per packet and the total number of resource elements), the Pareto front described above being tested using a range of intervals. Different Intentional Preference Values It is obtained by training an AI / ML model to optimize the adaptive parameters of the link.

[0162] Figure 15A flowchart according to some embodiments is shown. In some embodiments, the first network node 1500 may additionally: (1) acquire 1501 at least one training data sample associated with training an AI / ML model or algorithm for at least optimizing the communication configuration of a user device; and (2) train an AI / ML model or algorithm for at least optimizing the communication configuration of a user device based on the at least one training data sample, or send a fourth message 1503 to a second network node or a third network node 1506, the at least one training data sample associated with training an AI / ML model or algorithm for at least optimizing the communication configuration of a user device.

[0163] Therefore, in some embodiments, the first network node 1500 can be trained with an AI / ML model or algorithm to at least optimize the communication configuration of the user equipment. In another possible implementation of this method, such as... Figure 15 As shown, the second or third network node 1506 can train 1505 an AI / ML model or algorithm for at least optimizing the communication configuration of the user equipment used by the first network node. In this case, the first network node can additionally receive a fifth message 1507 from the second or third network node, which includes at least the AI / ML model or algorithm for at least optimizing the communication configuration of the user equipment. In some embodiments, the first network node 1500 can receive a sixth message 1509 from the second or third network node 1506 that includes a request for training data.

[0164] In some embodiments, a user equipment (UE) in a radio communication network may perform a method for optimizing communication with a first network node based on multiple intentions associated with the UE, the method comprising one or more of the following steps: (1) sending a first message to the first network node, the first message indicating information related to at least one intention for optimizing communication (i.e., sending and / or receiving) with at least one UE; and / or (2) receiving a third message from the first network node or a fourth message from a second network node, wherein the third message or the fourth message provides a communication configuration for the UE to optimize the communication with one or more intentions associated with the UE. The UE's method may additionally include the step of receiving a second message from the first network node, the second message requesting information related to at least one intention for optimizing communication (i.e., sending and / or receiving) with at least one UE. In one example, the UE may send the first message in response to receiving the second message.

[0165] In some embodiments, a second network node in a radio communication network performs a method for optimizing communication between a user equipment (UE) and a first network node based on multiple intentions associated with the UE. The method may include the step of sending a first message to the first network node, the first message indicating information related to at least one intention for optimizing communication (i.e., sending and / or receiving) with at least one UE. The method of the second network node may additionally include the step of receiving a second message from the first network node, the second message requesting information related to at least one intention for optimizing communication (i.e., sending and / or receiving) with at least one UE. In this case, the second network node may send the first message in response to receiving the second message. In one embodiment, the second network node may additionally: (i) receive a third message from the first network node, the third message providing a communication configuration for the UE to optimize one or more intentions associated with the UE; and / or (ii) send a fourth message to the UE, the fourth message providing the UE's communication configuration, or a portion thereof, obtained from the first network node.

[0166] In some embodiments, the first network node and / or the second network node may be different RAN nodes (e.g., two gNBs, or two eNBs, or two en-gNBs, or two ng-eNBs).

[0167] In some embodiments, the first network node and / or the second network node may be different nodes / functions of the same RAN node (e.g., gNB-CU-CP and gNB-DU, or gNB-CU-CP and gNB-CU-UP).

[0168] In some embodiments, the first network node may be a RAN node (e.g., gNB, eNB, en-gNB, or ng-eNB), and the second network node may be a component / node / function of the second RAN node (e.g., gNB-CU-CP).

[0169] In some embodiments, the first network node and / or the second network node may belong to the same radio access technology (e.g., E-UTRAN, NG-RAN, WiFi, etc.) or different radio access technologies (e.g., one belongs to NR, and the other belongs to E-UTRAN or WiFi).

[0170] In some embodiments, the first network node and / or the second network node may belong to the same RAN system (e.g., E-UTRAN, NG-RAN, WiFi, etc.) or different RAN systems (e.g., one belongs to NG-RAN and the other belongs to E-UTRAN).

[0171] In some embodiments, the first network node and the second network node may be connected via a direct signaling connection (e.g., two gNBs via XnAP) or an indirect signaling connection (e.g., e-NB and gNB via S1AP, NGAP and one or more core network nodes, such as MME and AMF).

[0172] In some embodiments, the first network node may be a management system, such as an OAM system or an SMO, while the second network node may consist of a RAN node or a function.

[0173] In some embodiments, the first network node may be a RAN node or a function, while the second network node may be a management system, such as OAM or SMO.

[0174] In some embodiments, the first network node may be a core network node or function, such as a 5GC function, while the second network node may consist of a RAN node or function.

[0175] In some embodiments, the first network node may be a RAN node or function, while the second network node may be a core network node or function, such as the 5GC function.

[0176] Figure 16 A method according to some embodiments is illustrated. In some embodiments, method 1600 is a computer-implemented method performed by a first network node in a communication network environment for optimizing transmission parameters of at least one user equipment.

[0177] Step s1601 of the method includes obtaining a first set of at least one intent of at least one user device.

[0178] Step s1603 of the method includes obtaining at least one intent preference value associated with at least one user device based on a first set of at least one intents acquired.

[0179] Step s1605 of the method includes using a first AI / ML model to determine the communication configuration of at least one user device based on at least one intent preference value.

[0180] Step s1607 of the method includes triggering a change in the configuration of at least one user equipment based on the communication configuration.

[0181] Figure 17 A method according to some embodiments is illustrated. Method 1700 is a computer-implemented method performed by a first user equipment in a communication network environment for optimizing the transmission parameters of the user equipment.

[0182] Step s1701 of the method includes generating a first message, which includes a first set of at least one intent of the user device.

[0183] Step s1703 of the method includes sending a first message to the first network node.

[0184] Step s1707 of the method includes obtaining the communication configuration.

[0185] Step s1707 of the method includes updating one or more transmission parameters based on the communication configuration.

[0186] Figure 18 This is a block diagram of an apparatus 1800 according to some embodiments. Apparatus 1800 may be a network node or a user equipment. Figure 18 As shown, the device may include: processing circuitry (PC) 1802, which may include one or more processors (P) 1855 (e.g., one or more general-purpose microprocessors and / or one or more other processors, such as application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), etc.); communication circuitry 1848, including a transmitter (Tx) 1845 and a receiver (Rx) 1847, for enabling the device to send and receive data via network 1810 (e.g., wirelessly sending / receiving data); and local storage unit (also referred to as a "data storage system") 1808, which may include one or more non-volatile storage devices and / or one or more volatile storage devices. In embodiments where PC 1802 includes a programmable processor, a computer program product (CPP) 1841 may be provided. CPP 1841 includes a computer-readable medium (CRM) 1842 for storing a computer program (CP) 1843 containing computer-readable instructions (CRI) 1844. CRM 1842 may be a non-transitory computer-readable medium, such as a magnetic medium (e.g., a hard disk), an optical medium, a storage device (e.g., random access memory, flash memory), etc. In some embodiments, CRI 1844 of computer program 1843 is configured such that, when executed by PC 1802, CRI causes the device to perform the steps described herein (e.g., the steps described herein with reference to the flowcharts). In other embodiments, the device may be configured to perform the steps described herein without code. That is, for example, PC 902 may consist of only one or more ASICs. Therefore, the features of the embodiments described herein can be implemented in hardware and / or software.

[0187] While various embodiments have been described herein, it should be understood that these embodiments are by way of example only and not as limiting examples. Therefore, the breadth and scope of this disclosure should not be limited to any of the embodiments described above. Furthermore, unless otherwise stated herein or the context clearly contradicts it, this disclosure covers any combination of the foregoing elements in all possible variations.

[0188] Furthermore, although the processes illustrated above and in the accompanying figures are presented as a series of steps, this is merely for illustrative purposes. Therefore, it is possible to add certain steps, omit certain steps, rearrange the order of steps, and execute certain steps in parallel.

[0189] abbreviation AMF 5G Access and Mobility Management Functions CN Core Network gNB-CU gNB Central Unit (CU) gNB-DU gNB Distributed Unit (DU)

Claims

1. A computer-implemented method (1600), performed by a first network node (600, 700, 1500, 1800) in a communication network environment, for optimizing transmission parameters for at least one user equipment (602, 702), the method comprising: Acquire (s1601) a first set of at least one intent of at least one user device; Based on the first set of at least one intents obtained, at least one intent preference value associated with the at least one user device is obtained (s1603); Using a first AI / ML model (902, 1002, 1102), the communication configuration of the at least one user device is determined (s1605) based on the at least one intent preference value; as well as Based on the communication configuration, a change in the configuration of the at least one user equipment is triggered (s1607).

2. The method according to claim 1, wherein, The first set of at least one intention includes one or more of the following: objective function Utility function Performance metrics Key performance indicators (KPIs) Quality of Service (QoS) requirements or Quality of Experience (QoE) requirements.

3. The method according to any one of claims 1 or 2, wherein, The objective function, utility function, performance metric, and KPI include one or more of the following: User throughput User spectrum efficiency, Each data packet transmits information, Resource utilization or allocation Delay, Block error rate Bit error rate reliability, Packet error rate Energy consumption Energy saving, Discontinuous transmission Discontinuous reception or Signal quality, and The QoS or QoE requirements include one or more of the following: Throughput requirements Spectral efficiency requirements Transmission delay requirements Block error rate requirements Packet error rate requirements Bit error rate requirement Packet error rate requirements Reliability requirements Signal-to-noise ratio requirements Resource utilization requirements Energy consumption requirements Energy saving requirements Resource utilization requirements Discontinuous transmission requirements Discontinuous reception requirements Transport block size requirement or Resource element requirements.

4. The method according to any one of claims 1 to 3, wherein, The communication configuration includes one or more of the following: The allocation or configuration of at least one communication resource The allocation or configuration of link adaptive parameters The allocation or configuration of information bits received or transmitted by the at least one user equipment. The allocation or configuration of downlink or uplink reference signals Configuration for discontinuous transmission or reception Or it can be used for energy-saving configurations.

5. The method according to any one of claims 1 to 4, wherein, The first set of at least one intents is contained in a first message (603, 703) sent by the at least one user equipment or the second network node.

6. The method according to claim 5, further comprising: Send a second message (605, 705) to the at least one user equipment or the second network node, the second message including a request for a first set of the at least one intent.

7. The method according to claim 5 or 6, further comprising: Send a third message (609, 709) to the at least one user equipment, the third message including the communication configuration.

8. The method according to any one of claims 1 to 7, wherein, Obtaining the at least one intent preference value includes: Map the first set of the at least one intent to the at least one intent preference value.

9. The method according to claim 8, wherein, The mapping includes: Map the first set of at least one intent to one or more user performance metrics parameterized relative to one or more intent preference values.

10. The method according to claim 9, wherein, The mapping is performed in the following manner: A user performance metric parameterized with respect to one or more intent preference values ​​is obtained by testing the first AI / ML model within a range of assumed values ​​for the one or more intent preference values.

11. The method according to any one of claims 8 to 10, wherein, The mapping is performed by a second AI / ML model.

12. The method according to any one of claims 1 to 10, wherein, The first set of at least one intent includes a first intent and a second intent, and wherein at least one intent preference value includes a first intent preference value associated with the first intent and a second intent preference value associated with the second intent.

13. The method according to claim 12, wherein, The first intent and the second intent correspond to the first user equipment.

14. The method according to claim 12, wherein, The first intent corresponds to a first user equipment, and the second intent corresponds to a second user equipment that is different from the first user equipment.

15. The method according to any one of claims 1 to 14, wherein, The first set of at least one intent includes at least one intent related to transmission reliability, throughput, spectral efficiency, block error rate, or delay.

16. The method according to any one of claims 12 to 15, wherein, The first intent corresponds to the number of information bits carried in the data packet transmission to or from the at least one user equipment. The second intent corresponds to the total number of resource elements used for communicating with the at least one user equipment up to the nth transmission attempt of the data packet, and The communication configuration includes one or more link adaptive parameters for downlink or uplink transmission.

17. The method according to any one of claims 1 to 16, wherein, Each component of the at least one intention r j Contained in an N-dimensional reward function vector r In, and wherein each component of the at least one intention preference value w j With the reward function vector r corresponding components r j Related.

18. The method of claim 17, wherein, Determining using the first AI / ML model includes providing the first AI / ML model with at least one of the following: A set of one or more information elements indicating or characterizing the state of the first network node, the state of the at least one user device, the state of the communication network environment related to communication between the first network node and the at least one user device, or a combination thereof. The indication of at least one intention, An indication of at least one intent preference value to be optimized for the at least one user device, or Its combination.

19. The method according to any one of claims 1 to 18, further comprising: Obtain at least one training data sample associated with the first AI / ML model or the second AI / ML model.

20. The method of claim 19, further comprising: The first AI / ML model is trained based on the at least one training data sample; or Send a fourth message (1503) to a third network node that includes the at least one training data sample.

21. The method of claim 20, further comprising: Receive a fifth message (1507), the fifth message including the first AI / ML model.

22. A node (600, 700, 1500, 1800) in a communication network environment for optimizing transmission parameters for at least one user equipment (602, 702), said node comprising: One or more memories (1842) include instruction data (1844) representing a set of instructions. as well as One or more processors (1855) are configured to communicate with the one or more memories and execute the set of instructions, wherein the set of instructions, when executed by the processors, causes the one or more processors to perform the method of any one of claims 1 to 21.

23. A computer program (1843) comprising instructions (1844) which, when executed by a processing circuit (1802), cause the processing circuit to perform the method of any one of claims 1 to 21.

24. A carrier comprising the computer program (1843) according to claim 23, wherein, The carrier includes one of electronic signals, optical signals, radio signals, or computer-readable storage media (1842).

25. A computer program product (1841) comprising a non-transitory computer-readable medium having the computer program of claim 23 stored thereon.

26. An apparatus (1800) comprising: Memory (1842); as well as A processing circuit (1802) coupled to the memory, wherein the apparatus is configured to perform the method of any one of claims 1 to 21.

27. A computer-implemented method (1700), performed by a first user equipment (602, 702, 1800) in a communication network environment, for optimizing transmission parameters for the user equipment, the method comprising: Generate (s1701) a first message (603, 703), the first message including a first set of at least one intent of the user equipment; Send the first message (s1703) to the first network node; Obtain (s1705) communication configuration; and Update one or more transmission parameters based on the communication configuration (s1707).

28. The method of claim 27, further comprising: Receive a second message (605, 705), the second message including a request for a first set of the at least one intent of the user equipment.

29. The method according to claim 27 or 28, wherein, The first set of at least one intention includes one or more of the following: objective function Utility function Performance metrics Key performance indicators (KPIs) Quality of Service (QoS) requirements or Quality of Experience (QoE) requirements.

30. The method according to any one of claims 27 to 29, wherein, The objective function, utility function, performance metric, and KPI include one or more of the following: User throughput User spectrum efficiency, Each data packet transmits information, Resource utilization or allocation Delay, Block error rate Bit error rate reliability, Packet error rate Energy consumption Energy saving, Discontinuous transmission Discontinuous reception or Signal quality, and The QoS or QoE requirements include one or more of the following: Throughput requirements Spectral efficiency requirements Transmission delay requirements Block error rate requirements Packet error rate requirements Bit error rate requirement Packet error rate requirements Reliability requirements Signal-to-noise ratio requirements Resource utilization requirements Energy consumption requirements Energy saving requirements Resource utilization requirements Discontinuous transmission requirements Discontinuous reception requirements Transport block size requirement or Resource element requirements.

31. The method according to any one of claims 27 to 30, wherein, The communication configuration includes one or more of the following: The allocation or configuration of at least one communication resource The allocation or configuration of link adaptive parameters The allocation or configuration of information bits received or transmitted by the at least one user equipment. The allocation or configuration of downlink or uplink reference signals Configuration for discontinuous transmission or reception Or it can be used for energy-saving configurations.

32. A user equipment (602, 702, 1800) in a communication network environment for optimizing transmission parameters of a user equipment, the user equipment comprising: One or more memories (1842) include instruction data (1844) representing a set of instructions. as well as One or more processors (1855) are configured to communicate with the one or more memories and execute the set of instructions, wherein, when executed by the processors, the set of instructions causes the one or more processors to perform the method of any one of claims 27 to 31.

33. A computer program (1843) comprising instructions (1844) which, when executed by a processing circuit (1802), cause the processing circuit to perform the method of any one of claims 27 to 31.

34. A carrier comprising the computer program (1843) according to claim 33, wherein, The carrier includes one of electronic signals, optical signals, radio signals, or computer-readable storage media (1842).

35. A computer program product (1841) comprising a non-transitory computer-readable medium having the computer program of claim 33 stored thereon.

36. An apparatus (1800) comprising: Memory (1842); as well as A processing circuit (1802) coupled to the memory, wherein the apparatus is configured to perform the method of any one of claims 27 to 31.

Images