AI / ML suitability reporting for PHY models

Abstract

A system and method for reporting suitability information of an artificial intelligence (AI) or machine learning (ML) model to a function is disclosed. In one embodiment, a method performed by a user equipment (UE) for reporting applicability information for at least one AI or ML model associated with a function includes transmitting applicability information for at least one AI or ML model associated with a function that has been configured and / or is being configured by the UE to a network node. In this manner, the UE can report such AI or ML model suitability information to the network in an improved manner compared to an existing UE capability reporting framework. Corresponding embodiments of the UE are also disclosed. Embodiments of a network node and a method of operating the same are also disclosed.

Description

Related applications

[0001] This application claims the benefit of provisional patent application serial number 63 / 494,670, filed April 6, 2023, the disclosure of which is incorporated herein by reference in its entirety. Technical Field

[0002] The present disclosure relates to a wireless communication system, and more particularly, to a user equipment (UE) reporting applicability information of at least one artificial intelligence (AI) / machine learning (ML) model associated with a function. Background Art

[0003] AI / ML in PHY Research Project Version 18

[0004] AI and ML have been studied by both academia and industry as promising tools for optimizing air interface design in wireless communication networks. Example use cases include: using autoencoders for channel state information (CSI) compression to reduce feedback overhead and improve channel prediction accuracy; using deep neural networks to classify line-of-sight (LOS) and non-line-of-sight (NLOS) conditions to improve positioning accuracy; and using reinforcement learning for beam selection on the network side and / or user equipment (UE) side to reduce signaling overhead and beam alignment latency; and using deep reinforcement learning to learn the optimal precoding strategy for complex multiple-input multiple-output (MIMO) precoding problems.

[0005] Within the 3rd Generation Partnership Project (3GPP) New Radio (NR) standardization work, a new Release 18 study project on AI / ML for the NR air interface was launched in May 2022. This study project will explore the benefits of enhancing the air interface's functionality to support AI / ML-based algorithms, thereby improving performance and / or reducing complexity / overhead. By examining several selected use cases (CSI feedback, beam management, and positioning), this study project aims to lay the foundation for future air interface use cases leveraging AI / ML technologies.

[0006] A possible high-level description of AI / ML model lifecycle management (LCM) for AI on the physical layer (PHY) could include Figure 1 The stages and data / signal flows are shown. A detailed description of the different LCM stages can be found in R1-2208908 “Discussion on general aspects of AI ML framework, Ericsson”.

[0007] Data collection is the phase where input data (raw data or preprocessed data) is collected and provided for model training, model inference, and model monitoring. AI / ML algorithm-specific data preparation (for example, data extraction and data refinement) is not performed during the data collection phase.

[0008] Model training is the process of training an AI / ML model using feature data from a training dataset and a validation dataset.

[0009] Model deployment is the process of converting AI / ML models into executable form and delivering it to the target UE that will perform model inference for inference.

[0010] Model inference is the process of using a deployed AI / ML model to generate a set of outputs based on a set of feature inputs.

[0011] Model monitoring is the process of monitoring drift in data and models or monitoring performance metrics after a model has been deployed. Based on the monitored performance, decisions such as model activation / deactivation / switching / fallback / selection can be made.

[0012] In 3GPP, the following terms are also assumed:

[0013] Include the following in the terminology working list for RAN1 AI / ML air interface study item (SI) discussions.

[0014] As research progresses, the description of terms can be further refined.

[0015] New terms may be added as research progresses.

[0016] Which subsets of terms will be included in the Technical Report (TR) are for further study (FFS).

[0017]

[0018]

[0019] The hypothesis at work

[0020]

[0021]

[0022] NOTE: Whether and how to indicate functionality will be discussed separately. Summary of the Invention

[0023] Disclosed are systems and methods for reporting applicability information of artificial intelligence (AI) or machine learning (ML) models for functions. In one embodiment, a method performed by a user equipment (UE) for reporting applicability information of at least one AI or ML model associated with a function includes sending, to a network node, applicability information of at least one AI or ML model associated with a function that has been configured and / or is being configured on the UE. In this manner, the UE can report such AI or ML model applicability information to the network in an improved manner compared to existing UE capability reporting frameworks.

[0024] In one embodiment, the applicability information of the at least one AI or ML model associated with the function includes any one or more of the following: an indication that the at least one AI or ML model associated with the function is not applicable, an indication that the function associated with the at least one AI or ML model is not applicable, a reason value, or a suggestion or recommendation or reconfiguration on how to make the at least one AI or ML model applicable.

[0025] In one embodiment, sending the applicability information includes sending a completion message, the completion message including the applicability information of the at least one AI or ML model associated with a function that has been configured and / or is being configured on the UE. In one embodiment, the completion message is a Radio Resource Control (RRC) Recovery Complete message, an RRC Setup Complete message, or an RRC Reconfiguration Complete message. In one embodiment, the method further includes: sending a request message to a network node; and receiving a response message from the network node in response to the request message, wherein sending the completion message including the applicability information includes: sending the completion message after receiving the response message. In one embodiment, the request message is an RRC Recovery Request, the response message is an RRC Recovery Message, and the completion message is an RRC Recovery Complete message. In another embodiment, the request message is an RRC Setup Request, the response message is an RRC Setup Message, and the completion message is an RRC Setup Complete message. In one embodiment, the response message includes configuration information for a function associated with the at least one AI or ML model, or information for activating a function associated with the at least one AI or ML model. In one embodiment, sending the completion message includes sending the completion message in response to configuration information of a function associated with the at least one AI or ML model, or information for activating a function associated with the at least one AI or ML model.

[0026] In one embodiment, the method further comprises receiving a reconfiguration message from the network node, wherein sending a completion message including the applicability information comprises sending a completion message after receiving the response message. In one embodiment, the reconfiguration message is an RRC reconfiguration message, and the completion message is an RRC reconfiguration complete message. In one embodiment, the reconfiguration message comprises configuration information of a function associated with the at least one AI or ML model, or information for activating a function associated with the at least one AI or ML model. In one embodiment, sending a completion message comprises sending a completion message in response to the configuration information of the function associated with the at least one AI or ML model, or information for activating a function associated with the at least one AI or ML model.

[0027] In one embodiment, sending the applicability information includes in or via a request message. In one embodiment, the request message is in an RRC recovery request.

[0028] Corresponding embodiments of a UE are also disclosed. In one embodiment, a UE for reporting applicability information of at least one AI or ML model associated with a function is adapted to send, to a network node, applicability information of at least one AI or ML model associated with a function that has been configured and / or is being configured on the UE.

[0029] In one embodiment, a UE for reporting applicability information of at least one AI or ML model associated with a function includes: a communication interface including a transmitter and a receiver; and processing circuitry associated with the communication interface. The processing circuitry is configured to cause the UE to transmit, to a network node, the applicability information of at least one AI or ML model associated with a function that has been configured and / or is being configured on the UE.

[0030] Also disclosed is an embodiment of a method performed by a network node. In one embodiment, a method performed by a network node for obtaining applicability information of at least one AI or ML model associated with a function includes receiving, from a UE, applicability information of at least one AI or ML model associated with a function that has been configured and / or is being configured on the UE.

[0031] In one embodiment, the applicability information of the at least one AI or ML model associated with the function includes any one or more of the following: an indication that the at least one AI or ML model associated with the function is not applicable, an indication that the function associated with the at least one AI or ML model is not applicable, a reason value, or a suggestion or recommendation or reconfiguration on how to make the at least one AI or ML model applicable.

[0032] In one embodiment, receiving the applicability information includes receiving a completion message including the applicability information of the at least one AI or ML model associated with the configured and / or configuring functions of the UE. In one embodiment, the completion message is one or more of the following: an RRC recovery complete message, an RRC setup complete message, and an RRC reconfiguration complete message. In one embodiment, the method further includes: receiving a request message from the UE; and sending a response message in response to the request message, wherein receiving the completion message including the applicability information includes: receiving the completion message after sending the response message. In one embodiment, the request message is an RRC recovery request, the response message is an RRC recovery message, and the completion message is an RRC recovery complete message. In another embodiment, the request message is an RRC setup request, the response message is an RRC setup message, and the completion message is an RRC setup complete message.

[0033] In one embodiment, the response message includes configuration information of a function associated with the at least one AI or ML model, or information for activating a function associated with the at least one AI or ML model. In one embodiment, receiving the completion message includes receiving the completion message in response to the configuration information of the function associated with the at least one AI or ML model, or information for activating a function associated with the at least one AI or ML model.

[0034] In one embodiment, the method further comprises sending a reconfiguration message to the UE, wherein receiving a completion message including the applicability information comprises: receiving the completion message after sending the response message. In one embodiment, the reconfiguration message is an RRC reconfiguration message, and the completion message is an RRC reconfiguration complete message. In one embodiment, the reconfiguration message comprises configuration information of a function associated with the at least one AI or ML model, or information for activating a function associated with the at least one AI or ML model. In one embodiment, receiving the completion message comprises: receiving the completion message in response to the configuration information of the function associated with the at least one AI or ML model, or information for activating a function associated with the at least one AI or ML model.

[0035] In one embodiment, sending the applicability information is included in a request message or is sent via a request message. In one embodiment, sending the request message is in an RRC recovery request.

[0036] In one embodiment, the method further includes performing one or more actions based on the suitability information.

[0037] Corresponding embodiments of a network node are also disclosed. In one embodiment, a network node for obtaining applicability information of at least one AI or ML model associated with a function is adapted to receive, from a UE, applicability information of at least one AI or ML model associated with a function that has been configured and / or is being configured on the UE.

[0038] In one embodiment, a network node for obtaining applicability information of at least one AI or ML model associated with a function includes a communication interface and processing circuitry associated with the communication interface. The processing circuitry is configured to cause the network node to receive, from a UE, applicability information of at least one AI or ML model associated with a function that has been configured and / or is being configured on the UE. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The accompanying drawings incorporated in and forming a part of this specification illustrate several aspects of the disclosure and, together with the description, serve to explain the principles of the disclosure.

[0040] Figure 1 A possible high-level description of AI / Machine Learning (ML) model lifecycle management (LCM) for artificial intelligence (AI) over the physical layer (PHY) is shown;

[0041] Figure 2 A process for a user equipment (UE) to report applicability information of at least one AI / ML model associated with a function in association with the UE transitioning from an inactive state to a connected state according to an embodiment of the present disclosure is illustrated;

[0042] Figure 3 A process in which a UE reports applicability information of at least one AI / ML model associated with a function in association with the UE transitioning from an inactive state to a connected state according to another embodiment of the present disclosure is shown;

[0043] Figure 4 A process is shown in which a UE reports applicability information of at least one AI / ML model associated with a function in association with the UE transitioning from an idle state to a connected state according to an embodiment of the present disclosure;

[0044] Figure 5 A process in which a UE reports applicability information of at least one AI / ML model associated with a function in association with the UE transitioning from an idle state to a connected state according to another embodiment of the present disclosure is shown;

[0045] Figure 6 An example of a communication system according to some embodiments is shown;

[0046] Figure 7 shows a UE according to some embodiments;

[0047] Figure 8 shows a network node according to some embodiments;

[0048] Figure 9 is a block diagram of a host computer according to various aspects described herein, which may be Figure 6 An embodiment of a host;

[0049] Figure 10 is a block diagram illustrating a virtualization environment in which functionality implemented by some embodiments may be virtualized; and

[0050] Figure 11 A communication diagram illustrating a host communicating with a UE via a network node over a partially wireless connection according to some embodiments. DETAILED DESCRIPTION

[0051] The embodiments set forth below provide information that will enable those skilled in the art to practice the embodiments and illustrate the best mode for practicing the embodiments. After reading the following description in light of the accompanying drawings, those skilled in the art will understand the concepts of the present disclosure and will recognize the applications of these concepts not specifically provided herein. It should be understood that these concepts and applications fall within the scope of the present disclosure.

[0052] There are currently certain challenges. The 3rd Generation Partnership Project (3GPP) has identified different phases of lifecycle management (LCM) for artificial intelligence (AI) or machine learning (ML) at the physical layer (PHY), as shown in R1-2208908. However, new issues have been identified in 3GPP TSG-RAN WG1 meeting #112, and the following is proposed in R1-2301868 "Final Summary of General Aspects of AIML Framework, Moderator (Qualcomm)":

[0053] To be able to develop models that are applicable to specific conditions (e.g., scenarios, configurations, sites, device types, etc.), investigate ways to associate datasets with specific applicability conditions:

[0054] - Auxiliary signaling of NW-side applicability information for UE-side data collection

[0055] - Auxiliary signaling of UE-side applicability information for NW-side data collection

[0056] Note: Research should consider the feasibility of disclosing proprietary information.

[0057] and

[0058] At least for the UE-side model and the UE part of the two-side model, RAN1 should study

[0059] - How to define and study the applicable conditions (sets) of a function / model.

[0060] oNote: Applicability conditions can be used to enable development of a scenario / configuration / [site] specific model and, if required, reporting on the applicability of that model to the network.

[0061] - Whether and how the UE reports supported functions (and, if necessary, supported models) and / or the applicable conditions(ies) for a supported set of functions.

[0062] - Whether and how performance requirements for features / models are defined (possibly as part of the applicable conditions)

[0063] - Potential enhancements to traditional UE features for reporting

[0064] In addition, RAN2 also reached agreement on the following CSI at RAN2#120 (Toulouse, November):

[0065] RAN2 scope includes procedures, protocols, and signaling for bilateral CSI use cases, such as:

[0066] 1. Ensure that the UE and gNB side models are configured / applied based on their applicable configurations / scenarios.

[0067] 2. Ensure that the models are correctly matched at both the UE and gNB sides, i.e., when a CSI encoder is used at the UE, the corresponding CSI decoder is used at the gNB.

[0068] 3. Realize simultaneous activation (deactivation) and switching of bilateral models

[0069] The problem to be solved by embodiments of the solution described herein is that AI / ML models for specific functions (e.g., beam management, channel state information (CSI), positioning) may or may not be applicable under certain conditions, but not all conditions. This is particularly true for user equipment (UE)-side AI / ML models.

[0070] For example, a UE capable of performing physical-layer AI / ML to implement beam management functionality may be equipped with an AI / ML model that has not been trained using a dataset associated with one or more beam configurations and / or one or more network areas (wide coverage, low-frequency layer). As a result, in some of these scenarios, the accuracy may not be appropriate and the AI / ML model may be deemed unsuitable or even unusable under certain conditions.

[0071] Even if the UE would report this in its UE capabilities and the fact that the UE is equipped with an AI / ML model for a given function / feature (e.g. CSI, beam management, positioning), there would still be various issues such as:

[0072] The applicability of the AI / ML model is dynamic and may depend on the location of the UE. The AI / ML model may not work in one cell, but as the UE moves, it may work in another cell to which the UE moves.

[0073] The applicability of AI / ML models is dynamic and may change after the UE performs AI / ML model training for new conditions.

[0074] The applicability of AI / ML models may only be effective under certain configurations of the UE configured by the network.

[0075] This dynamic nature of AI / ML model applicability makes it insufficient to be considered a typical UE capability.

[0076] Certain aspects of the present disclosure and its embodiments may provide solutions to these and other challenges. Disclosed is an embodiment of a method, performed by a UE, for reporting applicability information for at least one AI / ML model associated with a function. In one embodiment, the UE reports (e.g., in a completion message) the applicability information for at least one AI / ML model associated with a function that the UE has configured and / or is configuring. This provides the network with accurate information on the applicability of the UE-side AI / ML model function, i.e., whether the AI / ML model function supported by the UE can be used under certain conditions (e.g., in a given cell or a group of serving cells, under a given UE's current configuration, etc.).

[0077] Some example embodiments are as follows:

[0078] A1. A method at a user equipment (UE) for reporting applicability information of at least one AI / ML model associated with a function, the method comprising:

[0079] o Report (e.g., in a completion message) applicability information of at least one AI / ML model associated with a configured and / or configuring function of the UE.

[0080] A2. A method according to A1, wherein the report is in a completion message, and the completion message includes any one or more of the following:

[0081] oRRC establishment completed;

[0082] oRRC recovery completed;

[0083] oRRC reconfiguration completed.

[0084] A3. A method according to A1, wherein the applicability information of at least one AI / ML model associated with the function comprises one or more of the following:

[0085] o An indication that the AI / ML model associated with the feature is not applicable;

[0086] oReason value;

[0087] A4. The method of A1, wherein the UE reports applicability information of at least one AI / ML model associated with the function after it sends a request message, wherein the request message includes any one or more of the following:

[0088] oRRC recovery request message;

[0089] oRRC setup message;

[0090] A5. The method of A1, wherein the UE reports applicability information of the at least one AI / ML model associated with the function in response to receiving a message configuring the at least one AI / ML model associated with the function, wherein the message includes one or more of the following:

[0091] oRRC recovery message;

[0092] oRRC setup message;

[0093] oRRC reconfiguration message;

[0094] A6. A method at a user equipment (UE) for reporting applicability information of at least one AI / ML model associated with a function, the method comprising:

[0095] o Reporting, in or through a request message, applicability information of at least one AI / ML model associated with a function that has been configured and / or is being configured on the UE.

[0096] A7. A method according to A6, wherein the UE request message includes an RRC recovery request.

[0097] A8. The method according to A7, wherein the UE includes applicability information of at least one AI / ML model associated with the function in a message multiplexed with the RRC recovery request message.

[0098] Certain embodiments may provide one or more of the following technical advantages. One advantage of reporting an AI / ML model applicability indication to the network in a completion message, for example, during a state transition to the CONNECTED state, is that applicability is more dynamic than UE capabilities and may vary depending on, for example, the scenario, location, and UE configuration, which typically change during state transitions and / or mobility. Another advantage of such reporting is that it allows the UE to report its model requirements / conditions with greater granularity than can ultimately be conveyed using the existing UE capability reporting framework. Additionally, this reporting approach ultimately reduces the need to indicate unnecessary "applicability-related" information because it is associated with the current scenario and / or UE configuration.

[0099] 1 Initial Considerations

[0100] In this disclosure, the terms "ML model," "AI model," or "AI / ML model" are used interchangeably. An AI / ML model can be defined as a function or portion of a function deployed / implemented in a first node (e.g., a UE in the case of a UE-side model). An AI / ML model can be defined as a feature or portion of a feature implemented / supported in a first node. The first node can indicate the feature version to a second node. If the ML model is updated, the first node can change the feature version.

[0101] An AI / ML model may correspond to a function that receives one or more inputs (e.g., measurements, configurations) and provides as a result one or more specific types of predictions / estimations (e.g., time-domain and / or spatial-domain predictions of beam measurements). In one example, an ML model may correspond to a function that receives as input a measurement of a reference signal (e.g., transmitted in beam X) at time instance t0 and provides as a result a prediction of the reference signal at time t0+T. In another example, an ML model may correspond to a function that receives as input a measurement of reference signal X (e.g., transmitted in beam X) (e.g., a synchronization signal block (SSB) with index "x") and provides as a result a prediction of another reference signal transmitted in a different beam (e.g., reference signal Y (e.g., transmitted in beam X) with index "x"). Another example is an ML model used to assist in CSI estimation. In this setting, the ML model is specific to both the UE-side ML model and the network (NW)-side ML model. These two ML models jointly provide a joint network. The function of the ML model at the UE is to compress the channel input, while the function of the ML model on the NW side is to decompress the output received from the UE. A similar approach can also be applied to positioning, where the input can be some form of channel impulse temporally related to a reference point (usually the TP (transmitting point)). The goal on the NW side is to detect distinct peaks in the impulse response, which reflect the multipath traversed by the radio signal to reach the UE. Another positioning approach is to input multiple sets of measurements into the ML network and derive an estimated UE position based on these. Another ML model can assist the UE with channel or interference estimation. Channel estimation can, for example, be for the Physical Downlink Shared Channel (PDSCH) and associated with a specific set of reference signal patterns sent from the NW to the UE. The ML model will then be part of the receive chain within the UE and may not be directly visible in the reference signal patterns configured / scheduled between the NW and the UE. Another example of an ML model for CSI estimation is to predict a suitable channel quality indicator (CQI), precoding matrix indicator (PMI), rank indicator (RI), CSI-RS resource indicator (CRI), or similar value in the future. The future can be a certain number of time slots after the UE has performed the last measurement, or for a specific time slot in the future.

[0102] In this disclosure, the term "beam" may correspond to a spatial direction in which a signal is transmitted (e.g., by a network node) or received (e.g., by a UE), or a spatial filter applied to a transmitted or received signal. Thus, transmitting a signal using different beams may correspond to transmitting the signal in different spatial directions. When reference is made herein to a "selected beam," it may refer to a beam index and / or a reference signal (RS) index or identifier, such as an SSB index or a CSI reference signal (CSI-RS) resource identifier. Thus, selecting a beam may correspond to selecting an SSB associated with an SSB index, or selecting a beam may correspond to selecting a CSI-RS associated with a CSI-RS resource identifier.

[0103] The network (NW) in the present disclosure may be a general NW node, a gNodeB (gNB), a base station, a unit within a base station for processing at least some ML operations, a relay node, a core network node, a core network node processing at least some ML operations, a device supporting device-to-device (D2D) communication, a location management function (LMF), or other types of location servers.

[0104] Another way to describe AI / ML models is as follows:

[0105] In terms of the time / frequency / space domain, the output of the AI / ML model can be at a different time instance, frequency location, spatial orientation, or a combination of time / frequency / space than the model input. In one example (time domain), the ML model can correspond to a function that receives as input the measurement of a reference signal (e.g., transmitted in beam X) at time instance t0 and provides as a result a prediction of the reference signal at time instance t0+T. In another example (spatial domain), the ML model can correspond to a function that receives as input the measurement of reference signal X (e.g., the SSB with index "x") (e.g., transmitted in beam x) and provides as a result an estimate / prediction of the link quality of another reference signal (e.g., reference signal Y) transmitted in a different beam (e.g., transmitted in beam y).

[0106] In terms of model structure, the ML model can be fully contained in the UE or split between the UE and the network.

[0107] An example of a segmented structure is an ML model used to assist in CSI estimation, where one possible configuration of the ML model is a segmented model that includes a specific sub-ML model within the UE and a sub-ML model within the NW side, which collaborate to generate the desired results for the overall ML model. The function of the sub-ML model at the UE is to compress the channel input, and the function of the sub-ML model on the NW side is to decompress the output received from the UE. A similar approach can also be applied to positioning, where the input can be some form of channel impulse related to a reference point in time. The goal on the NW side is to detect different peaks in the impulse response, which correspond to different reception directions of the radio signal on the UE side.

[0108] One example of ML included within the UE is ML-enhanced positioning, e.g., an ML model implemented in the UE takes as input multiple sets of measurements (each set corresponding to downlink (DL) signals from a different network node) and, based on this, derives an estimated position of the UE.

[0109] In terms of practicality at the physical layer, ML models can be used for a number of functions, including channel estimation, line-of-sight (LOS) / non-line-of-sight (NLOS) classification, beam selection, UE position estimation, link adaptation, and more. For example, an ML model can assist the UE with channel estimation, which may or may not include interference estimation. The channel estimate can, for example, be for the Physical Downlink Shared Channel (PDSCH) and associated with a specific set of reference signal patterns sent from the NW to the UE. The ML model will then be part of the receive chain within the UE and may not be directly visible in the reference signal patterns configured / scheduled for use between the NW and the UE. Another example of an ML model used for channel state information (CSI) estimation is predicting appropriate CQI, PMI, RI, or similar values ​​in the future. This future can be a certain number of time slots after the UE has last performed a measurement, or for a specific time slot in the future.

[0110] According to an embodiment of the present disclosure, a UE is connected to a network (e.g., it can receive and transmit data and / or control information), i.e., is in the RRC_CONNECTED state, and is configured to perform specific functions (which may be referred to as AI / ML model functions, such as beam measurement prediction in the time domain) by using an AI / ML model. The specific functions of the AI / ML model may be used, for example, for one of the following examples, which may also be grouped into a functional area (one or more AI / ML model functions per area) as shown below:

[0111] CSI Report

[0112] Beam Management (BM)

[0113] o In one option, there may be a BM function for the AI / ML model, where the AI / ML model (e.g., at the UE) is capable of performing inference on one or more time-domain predictions related to beam management. For example, the network may configure the UE to report one or more time-domain predictions of SSB and / or CSI-RS and / or Phase Tracking Reference Signal (PTRS) measurements (e.g., on the Physical Uplink Control Channel (PUCCH) and / or Physical Uplink Shared Channel (PUSCH), for example, by receiving a reporting configuration for the AI / ML.

[0114] o In one option, there may be a BM functionality of an AI / ML model, where the AI / ML model (e.g., at the UE) is able to perform inference of one or more spatial domain predictions related to beam management.

[0115] o In one option, there may be a BM functionality of an AI / ML model, where the AI / ML model (e.g., at the UE) is able to perform reasoning for both time domain prediction and spatial domain prediction related to beam management.

[0116] o A UE is considered to be configured with AI / ML functionality when at least one action associated with the functionality is configured, e.g. the UE is configured to report a prediction of beam measurements to one of its configured serving cells, and / or report the CSI and / or SSB of the serving cell.

[0117] Radio Resource Management (RRM) measurements

[0118] o For example, mobility measurements, i.e., Reference Signal Received Power (RSRP), Reference Signal Received Quality (RSRQ), Received Signal Strength Indicator (RSSI), but also aspects related to radio link failure, such as Radio Link Failure (RLF) prediction. Furthermore, predictions related to the timer (T310) and counters (N310 and N311) related to Radio Link Monitoring (RLM) can also be considered here.

[0119] o The measurement framework, such as defined in §5.5, 3GPP Technical Specification (TS) 38.331 (e.g., see V16.12.0 and V17.4.0), includes how the UE performs measurements (e.g., measurement configuration), what triggers measurement reporting (e.g., event-triggered reporting, periodic reporting), and what to include in the measurement report.

[0120] Link adaptation

[0121] Hybrid Automatic Repeat Request (HARQ) transmission

[0122] Data transmission

[0123] Data reception

[0124] Power control

[0125] UE positioning

[0126] Random access transmission

[0127] Energy efficiency (e.g., discontinuous reception (DRX) settings)

[0128] This disclosure refers to the applicability of an AI / ML model to a function, which refers to the property of the AI / ML model for that function that can be configured and used by the UE (e.g., by the UE in the case of a UE-side model) under certain conditions. The AI / ML for a function is considered applicable when the AI / ML model is able to:

[0129] - Producing outputs, such as time-domain predictions of beam measurements and / or when the AI / ML model is able to produce outputs (e.g., time-domain predictions of beam measurements) with suitable (e.g., good enough) accuracy. For BM, accuracy may include, for example:

[0130] o Average Layer 1 (L1) Reference Signal Received Power (RSRP) difference between the best (Top-1) predicted beam and the ideal beam (scanning all beams)

[0131] o Cumulative distribution function (CDF) percentile of the L1-RSRP difference of the best predicted beam

[0132] o Beam prediction accuracy (%), with a margin of 1 decibel (dB) for the best beam.

[0133] -The UE may also receive specific thresholds from the network (NW) to compare with accuracy key performance indicators (KPIs) to understand whether the model is “fit for purpose”.

[0134] - Produces an output that reveals some confidence value for the prediction. If the UE is able to estimate the confidence of each prediction, then the W can decide on a per-input basis whether to use that prediction. That is, the model is "fit" for some of its empirical data. For example, the UE may report

[0135] o The predicted L1-RSRP / Signal to Interference and Noise Ratio (SINR) of the beam (DL and / or Uplink (UL)) with probability x and the predicted confidence interval within which it lies. For example, the probability that the L1-RSRP SINR is within the range [8dB, 10dB] is 95%.

[0136] o The predicted value can be reported as a probability density function using, for example, a Gaussian mixture, and then reported with parameters describing the components of the Gaussian mixture: its mean, variance, and the component weights for each component.

[0137] - The model has been collected and trained in the current UE configuration / scenario, where the model is no older than a certain threshold T. The NW can determine T based on deployment changes (new beam pattern, new cell, etc.).

[0138] When the AI / ML model cannot produce an output (e.g., a time-domain prediction of beam measurement) and / or when the AI / ML model can produce an output (e.g., a time-domain prediction of beam measurement) but with inappropriate accuracy (e.g., insufficient accuracy), the AI / ML model for a function is considered unsuitable. In the case where a UE has multiple AI / ML models for the same function (where different models are applicable to different scenarios), the AI / ML model for that function is considered unsuitable when all AI / ML models for that function are unsuitable (otherwise, the UE does not report that the model is unsuitable but instead switches to another applicable model for that function). Similarly, when the UE is configured, the AI / ML model for that function is considered applicable when at least one AI / ML model for that function is applicable.

[0139] According to this disclosure, there are different reasons why AI / ML models may not be applicable, such as:

[0140] - Location / geographic area

[0141] o The AI / ML model for a function is applicable in a first area of ​​the network to which the UE is registered, but the same AI / ML model for the function may not be applicable in a second area of ​​the network to which the UE is registered. For example, when a UE camps on cell A in idle mode and transitions to connected state, the AI / ML model for the function is applicable to cell A. However, if the UE in idle state performs cell reselection and moves to cell C, the AI / ML model for the function may not be applicable.

[0142] oOne reason for this could be that the training dataset used to train the AI / ML model may be representative of one region but not another.

[0143] o Another reason could be due to regulatory restrictions, for example, a functional AI / ML model can only be used in certain locations.

[0144] o“Location” in this context may include one or more of the following:

[0145] ■ One or more cells, for example, one or more cells defined by one or more cell identifiers such as a global cell identifier.

[0146] ■ One or more tracking areas

[0147] ■ One or more tracking area codes

[0148] ■One or more registration areas

[0149] ■ One or more notification areas based on the Radio Access Network (RAN)

[0150] ■ GPS location and / or demarcated area

[0151] ■ Within the coverage area of ​​the Wireless Local Area Network (WLAN) Access Point (AP) list

[0152] ■ Within the coverage area of ​​the Bluetooth beacon list

[0153] ■ Deployment type, such as small cell, large cell, indoor, outdoor

[0154] ■ A list of Public Land Mobile Network (PLMN) and / or Non-Public Network (NPN) identifiers (for example, an NPN identifier may indicate a specific factory setup, and any training data used in that factory may only be applicable within that factory and not elsewhere)

[0155] -UE configuration

[0156] o When the UE is configured with a first configuration (including lower layers, bearer configuration, measurement configuration, etc.), such as represented by RRCReconfiguration (1), the AI / ML model of the function may be applicable. However, when the UE is configured with a second configuration (including lower layers, bearer configuration, measurement configuration, etc.), such as represented by RRCReconfiguration (2), the same AI / ML model of the function may not be applicable. For example, when the UE transitions to the connected state and receives a configuration equivalent to RRCReconfiguration (1), the AI / ML model of the function is applicable. However, if the UE transitions to the connected state and receives a configuration equivalent to RRCReconfiguration (2), the AI / ML model of the function is deemed not to be applicable.

[0157] o One reason for this is that for a given UE configuration, the training dataset used to train the AI / ML model may result in the model not producing accurate output (in inference) for some UE configurations; for example, the AI / ML model for that function is suitable for predicting measurements in a first set of frequencies (e.g., frequency range 1 (FR1) and / or f0, f1, f2), but not for predicting measurements in a second set of frequencies (e.g., FR2 and / or f7, f8, f9).

[0158] Another reason could be that the AI / ML model was trained using a certain CSI-RS periodicity. For example, the UE expects a 20ms periodicity to be able to perform channel prediction for the next 10ms (between measurements). However, the UE may be configured with aperiodic CSI-RS or a 40ms periodicity, making the model inaccurate.

[0159] o Another reason could be that the new configuration may not be applicable to existing configuration parameters already set in the UE, for example, due to conflicts with non-AI / ML related functions, or related to hardware / software limitations.

[0160] -Network configuration

[0161] oSingle beam and multi-beam

[0162] oConfiguration information broadcast by the network

[0163] oBeamforming mode

[0164] o For example, the NW indicates that it uses AI / ML to perform beam prediction, CSI prediction, or positioning, so the UE should not activate this feature

[0165] -Mobile features

[0166] o When a UE's mobility characteristics fall into one category, the AI / ML model for that function may be applicable, but the same AI / ML model for that function may not be applicable to a second category of mobility characteristics. For example, a UE may be classified as slow (classified by the UE based on its sensor-based measurements and / or network-defined criteria (e.g., speedStateReselectionPars defined in RRC specification TS 38.331 v17.3)), and the training data used to train the AI-ML model belongs exclusively to that mobility category. However, if the UE falls into a different mobility category, the AI / ML model for that function may not be applicable when transitioning from an idle / inactive state.

[0167] ■ For example, time beam prediction may be applicable based on UE mobility only, such as when the UE is moving at a constant / near-constant speed. Or based on the type of mobility, such as in a vehicle or train, which can provide more predictable trajectories. Or based on whether the UE is rotating.

[0168] How to determine if an AI / ML model is suitable for use?

[0169] You can use the following methods to determine whether the model is suitable.

[0170] · UE-based methods : For example, the UE determines whether the model is applicable by comparing the experienced configuration, location and / or mobility with the training data used by the model.

[0171] · NW-assisted methods For example, the NW can always activate the UE-side model. Based on model monitoring, the NW can check the model's KPIs and performance. The NW can signal to the UE that the model is not suitable for the current configuration, and the UE can then update the model's suitability. In another embodiment, the NW can also indicate certain requirements for the model to the UE. For example, the NW requires the model to provide a reliable confidence metric for its predictions, or, as another example, the required accuracy of the model.

[0172] Inactive

[0173] In one option, the Inactive state corresponds to a protocol state in which the UE stores the UE context and considers the connection (e.g., one or more bearers) to be suspended. When the UE attempts to resume the connection and transitions to the Connected state, it restores the UE context. When the UE determines to resume the connection, it sends a Resume Request message (e.g., RRCResumeRequest or RRCResumeRequest1) and may receive an RRC Resume message in response, according to which the UE enters the Connected state. In this case, the UE Inactive AS Context may correspond to the UE context stored when the connection was suspended and restored when the connection is resumed.

[0174] 2. Examples related to transitioning from an inactive state

[0175] like Figure 2 As shown in the example of , in one set of embodiments, a UE in an inactive state (e.g., RRC_INACTIVE) sends a request message (e.g., an RRC recovery request message, such as RRCResumeRequest or RRCResumeRequest1 defined in 3GPP Technical Specification (TS) 38.331) to a network (e.g., a gNodeB) (step 200), and receives a response message (e.g., RRC recovery, such as RRCResume message defined in 3GPP TS 38.331) (step 208), based on which the UE enters a connected state (RRC_CONNECTED), wherein the response message (e.g., RRCResume) configures (and / or activates) or attempts to configure one or more AI / ML model functions for the UE.

[0176] -For example, one AI / ML model function may correspond to one or more time domain predictions of a UE being configured to report and / or perform SSB and / or CSI-RS measurements (e.g., to one of its configured serving cells) (e.g., of resources associated with one of its configured serving cells).

[0177] -For example, an AI / ML model function may correspond to one or more spatial domain predictions that a UE is configured to report and / or perform (e.g., to one of its configured serving cells) SSB and / or CSI-RS measurements (e.g., of resources associated with one of its configured serving cells).

[0178] - For example, an AI / ML function may correspond to a UE being configured to report and / or perform CSI prediction;

[0179] - For example, an AI / ML function may correspond to the UE being configured to report and / or perform positioning prediction.

[0180] According to this approach, an alternative would be for the NW to include an explicit configuration in its response message (e.g., in step 208), which further allows the UE to report applicability-related information of the AI / ML model function in the completion message.

[0181] In another alternative solution, the UE is allowed to proactively report the applicability-related information of the AI / ML model function in the completion message (step 212).

[0182] In response to one or more AI / ML model functions being configured (e.g., in step 208), the UE sends a completion message (e.g., an RRC recovery complete message) to the network, the completion message including at least one indication indicating that at least one of the configured (or to be configured / to be activated) AI / ML model functions being configured (and / or activated) is not applicable (when the UE determines that the AI / ML model is not applicable) (step 212). According to the method, in one option, before the UE sends a completion message (e.g., an RRC recovery complete message) to the network (the completion message including at least one indication indicating that at least one of the configured (or to be configured / to be activated) AI / ML model functions being configured (and / or activated) is not applicable (when the UE determines that the AI / ML model is not applicable)) (e.g., in step 212), the UE determines whether the AI / ML model function is applicable (step 210).

[0183] According to this method, in one option, when the UE sends an RRC resume request message to the target network node (e.g., target gNodeB) (e.g., in step 200), the UE is not configured with the AI / ML model function: that is, the UE does not have the configuration of the AI / ML model function in its stored UE context (UE access stratum inactive context). In other words, the AI / ML model function configuration (e.g., reporting configuration for the UE to report one or more time-domain predictions of beam measurements (e.g., synchronization signal (SS)-reference signal received power (RSRP) prediction of the serving cell)) is explicitly included in the radio resource control (RRC) resume message (e.g., in step 208). In response to this configuration, the UE determines whether the AI / ML model function is applicable (e.g., in step 210).

[0184] According to this method, in another option, when the UE sends an RRC resume request message to the target network node (e.g., target gNodeB) (e.g., in step 200), the UE is configured with the AI / ML model function: that is, the UE has the configuration of the AI / ML model function in its stored UE context (UE access stratum inactive context). When the UE receives the RRC resume message (e.g., in step 208), the AI / ML model function is resumed, and the UE determines whether the resumed AI / ML model function is applicable under the configuration resulting from the UE applying the RRC resume message (e.g., in step 210).

[0185] According to the method, in one option, when the UE determines that the AI / ML model function is not applicable (e.g., in step 210), the UE includes the indication in the RRC recovery complete message and sends the RRC recovery complete message to the target network node (e.g., in step 212), wherein the indication is indicating that the configured (or to be configured / to be activated) AI / ML model function whose configuration is included in the RRC recovery is not applicable (e.g., under the current scenario and / or configuration).

[0186] According to the method, in one option, when the UE determines that the AI / ML model function is applicable (e.g., in step 210), the UE does not include the indication in the RRC recovery complete message and sends the RRC recovery complete message to the target network node (e.g., in step 212). When the target network node receives the RRC recovery complete message without the indication, the target network node considers that the configured AI / ML model function is applicable.

[0187] like Figure 3As shown in the example of , in one set of embodiments, the UE is in an inactive state (e.g., RRC_INACTIVE), and when the UE sends an RRC resume request message to a target network node (e.g., a target gNodeB) (e.g., in step 304), the UE is configured with at least one AI / ML model function: that is, the UE has the configuration of at least one AI / ML model function in its stored UE context (UE access stratum inactive context). When the UE needs to send an RRC resume request (or during resumption preparation), the UE resumes the AI / ML model function (step 300), and the UE determines whether the resumed AI / ML model function is applicable under the current UE configuration (already resumed) and existing conditions (e.g., the target cell to which the UE is attempting to resume) (step 302).

[0188] -In one option, when the UE determines that the AI / ML model functionality is not applicable in step 302, the UE includes the indication in an RRC recovery request message, and in step 304, sends an RRC recovery request message to the target network node, wherein the indication is indicating that the restored AI / ML model functionality in the UE context is not applicable (e.g., under the current scenario and / or configuration). The RRC recovery request enables the target network node to retrieve the UE context, and after receiving the UE context and having previously received the indication in the RRC recovery request, the target network node may release or deactivate the AI / ML functionality that has been reported as not applicable in the RRC recovery request.

[0189] - In another option, when the UE determines that the AI / ML model function is not applicable in step 302, the UE includes the indication in a message to be multiplexed with the RRC recovery request message, and in step 304, sends an RRC recovery request message to the target network node, wherein the indication Instructing The resumed AI / ML model functionality in the UE context is not applicable (e.g., under the current scenario and / or configuration). The RRC resume request enables the target network to retrieve the UE context and, upon receipt of the UE context and having previously received this indication in the RRC resume request, the target network node may release or deactivate the AI / ML functionality that had been reported as not applicable in the RRC resume request.

[0190] In another option, when the UE determines that the AI / ML model functionality is not applicable in step 302, the UE deactivates it when resuming the AI / ML model functionality. This operation can be combined with reporting the indication of inapplicability in the RRC resume request in step 304. The benefit here is that if the target network node does not want the inapplicable AI / ML model, it will not have to have it because it will be deactivated by the UE anyway.

[0191] In another option, when the UE determines that the AI / ML model function is not applicable in step 302, the UE releases the configuration of the AI / ML model function. This operation can be combined with reporting the indication of inapplicability in the RRC recovery request in step 304. The benefit here is that if the target network node does not want the unapplicable AI / ML model, it will not have to have it because it will be released by the UE.

[0192] According to the method, in one option, when the UE determines that the AI / ML model functionality is applicable in step 302, the UE does not include the indication in the RRC resume request message in step 304. The UE sends the RRC resume request message to the target network node. When the target network node receives the RRC resume request message without the indication, the target network node considers that the restored AI / ML model functionality is applicable.

[0193] 3. Embodiments Related to Transition from Idle State

[0194] Figure 4 An example of a set of embodiments is shown, in which a UE in an idle state (e.g., RRC_IDLE) sends a request message (e.g., an RRC setup request message, such as RRCSetupRequest defined in 3GPP TS 38.331) to a network (e.g., a gNodeB) (step 400), and receives a response message (e.g., an RRC setup message, such as RRCSetup message defined in TS 38.331) (step 404), based on which the UE enters a connected state (RRC_CONNECTED), wherein the response message (e.g., RRCSetup) configures (and / or activates) or attempts to configure one or more AI / ML model functions for the UE.

[0195] -In one option, one or more AI / ML model capabilities may be configured to the UE before security is activated.

[0196] - In one option, the target network node (e.g., the target gNodeB to which the UE is attempting to transition to the connected state) is able to retrieve the UE capabilities regarding the AI / ML model functionality based on information about the UE in the RRC setup request (e.g., the UE identifier or a portion thereof), such as bits of the UE identifier, e.g., bits of the 5GS-TMSI assigned when the UE was registered, if the UE is registered.

[0197] -For example, one AI / ML model function may correspond to one or more time domain predictions of a UE being configured to report and / or perform SSB and / or CSI-RS measurements (e.g., to one of its configured serving cells) (e.g., of resources associated with one of its configured serving cells).

[0198] -For example, an AI / ML model function may correspond to one or more spatial domain predictions that a UE is configured to report and / or perform (e.g., to one of its configured serving cells) SSB and / or CSI-RS measurements (e.g., of resources associated with one of its configured serving cells).

[0199] - For example, an AI / ML function may correspond to a UE being configured to report and / or perform CSI prediction;

[0200] - For example, an AI / ML function may correspond to the UE being configured to report and / or perform positioning prediction.

[0201] According to this method, an alternative would be that in step 408, the NW includes an explicit configuration in its response message (e.g., in step 404), which further allows the UE to report applicability-related information of the AI / ML model function in the completion message.

[0202] In another alternative solution, the UE is allowed to proactively report the applicability-related information of the AI / ML model function in the completion message.

[0203] In response to one or more AI / ML model functions being configured in the RRC setup message, the UE sends a completion message (e.g., an RRC setup complete message) to the network, the completion message including at least one indication indicating that at least one of the configured (or to be configured / activated) AI / ML model functions being configured (and / or activated) is not applicable (when the UE determines that the AI / ML model is not applicable) (step 408).

[0204] According to the method, in one option, in step 408, before the UE sends a completion message (e.g., an RRC setup complete message) to the network (the completion message including at least one indication indicating that at least one of the configured (or to be configured / activated) AI / ML model functions being configured (and / or activated) is not applicable (when the UE determines that the AI / ML model is not applicable)), the UE determines whether the AI / ML model function is applicable (step 406).

[0205] According to the method, in one option, when the target network node receives the RRC setup complete message in step 408, the target network node begins to know the UE's capabilities in terms of the AI / ML model function (these capabilities can be retrieved from the core network) and its applicable status (e.g., based on the indication or the absence of the indication) for the UE's current configuration (e.g., according to RRC setup) and conditions (e.g., radio environment, target cell and / or area to which the UE is connected). Therefore, the target network node is able to determine whether to configure (or activate) the AI / ML model function in the RRC reconfiguration that also configures the data radio bearer for the UE that has just transitioned to RRC_CONNECTED.

[0206] like Figure 5 As shown in the example of , in one set of embodiments, a UE in an idle state (e.g., RRC_IDLE) sends a request message (e.g., an RRC setup request message, such as RRCSetupRequest defined in TS 38.331) to a network (e.g., a gNodeB) (step 500), receives a response message (e.g., RRC setup, such as RRCSetup message defined in TS 38.331), based on which the UE enters a connected state (RRC_CONNECTED) (step 504), activates access stratum (AS) security (see, e.g., 506-522), and receives an RRC reconfiguration message (e.g., RRCReconfiguration), which configures (and / or activates) or attempts to configure one or more AI / ML model functions for the UE (step 524).

[0207] -In one option, one or more AI / ML model capabilities may be configured to the UE after security is activated.

[0208] -In one option, when the UE context is established at the target network node (e.g., when the target gNodeB receives the initial UE context establishment request), the target network node (e.g., the target gNodeB where the UE is attempting to transition to the connected state) can retrieve the UE capabilities regarding the AI / ML model functions.

[0209] -For example, one AI / ML model function may correspond to one or more time domain predictions of a UE being configured to report and / or perform SSB and / or CSI-RS measurements (e.g., to one of its configured serving cells) (e.g., of resources associated with one of its configured serving cells).

[0210] -For example, an AI / ML model function may correspond to one or more spatial domain predictions that a UE is configured to report and / or perform (e.g., to one of its configured serving cells) SSB and / or CSI-RS measurements (e.g., of resources associated with one of its configured serving cells).

[0211] - For example, an AI / ML function may correspond to a UE being configured to report and / or perform CSI prediction;

[0212] - For example, an AI / ML function may correspond to the UE being configured to report and / or perform positioning prediction.

[0213] In response to one or more AI / ML model functions being configured in the first RRC reconfiguration message when the UE transitions to the connected state, the UE sends a completion message (e.g., an RRC reconfiguration complete message) to the network, the completion message including at least one indication indicating that at least one of the configured (or to be configured / to be activated) AI / ML model functions being configured (and / or activated) is not applicable (when the UE determines that the AI / ML model is not applicable) (step 528).

[0214] According to the method, in one option, in step 528, before the UE sends a completion message (e.g., an RRC reconfiguration complete message) to the network (the completion message including at least one indication indicating that at least one of the configured (or to be configured / activated) AI / ML model functions being configured (and / or activated) is not applicable (when the UE determines that the AI / ML model is not applicable)), the UE determines whether the AI / ML model function is applicable (step 526).

[0215] A set of embodiments in Section 3 describes a scenario in which a UE is transitioning to an idle state and receives a first RRC reconfiguration that includes configuration of at least one AI / ML model function, for example, the UE is configured to report and / or perform one or more time / spatial predictions of beam measurements. However, these methods are also applicable to at least the case in which the UE is in an RRC_CONNECTED state and receives any RRC reconfiguration that configures (e.g., adds and / or modifies) an AI / ML model function, such that an indication of the applicability of the AI / ML model function being configured and / or added and / or modified can be included in an RRC reconfiguration completion that is not necessarily the first RRC reconfiguration completion after the UE transitions from the idle state.

[0216] 4. Indications of AI / ML model functionality not being applicable

[0217] According to one embodiment, the UE reports the applicability information of at least one AI / ML model associated with the function. According to one embodiment, the UE reports the applicability information of at least one AI / ML model associated with the function that the UE has configured and / or is configuring in a completion message.

[0218] In the above sections 2 and 3, the applicability information of at least one AI / ML model associated with a function is represented as an indication that the AI / ML model associated with the function is not applicable. However, this should not be limited to this, and in this section, in addition to and / or instead of the indications in sections 2 and 3, and / or instead of the indications in sections 2 and 3, other examples of applicability information of AI / ML models associated with functions that can be reported are also presented. In this article, the terms "indication" or applicability information are used interchangeably.

[0219] Therefore, according to one embodiment, the applicability information reported by the UE (or an indication that the AI / ML model function is not applicable) includes any one or more of the following:

[0220] -1) An indication that the AI / ML model functionality that the UE has been configured with or is being configured with is not applicable.

[0221] o Alternatively, an indication that the UE has been configured or is being configured with an AI / ML model function that is applicable. The absence of this indication indicates that the AI / ML model function is not applicable.

[0222] -2) Multiple indications that multiple AI / ML model functions that the UE has been configured with or is being configured with are not applicable.

[0223] oIn this case, the UE may have multiple AI / ML model capabilities, e.g.

[0224] ■AI / ML Model Functionality 1) BM functionality of the AI / ML model, where the AI / ML model (e.g., at the UE) is capable of performing inference on one or more time-domain predictions related to beam management. For example, the network may configure the UE to report one or more time-domain predictions of SSB and / or CSI-RS measurements (e.g., on PUSCH and / or Physical Uplink Control Channel (PUCCH)), for example, by receiving a reporting configuration for the AI / ML.

[0225] ■AI / ML model functionality 2) BM functionality of the AI / ML model, where the AI / ML model (e.g., at the UE) is capable of performing inference on one or more spatial domain predictions related to beam management.

[0226] ■AI / ML model functionality 3) BM functionality of the AI / ML model, where the AI / ML model (e.g., at the UE) is able to perform reasoning for both time domain prediction and spatial domain prediction related to beam management.

[0227] o The UE may then be configured with these multiple AI / ML model capabilities a), b), c), such as all of these different reports described in a), b), c).

[0228] o The UE then determines for each one whether the functionality is applicable and includes an indication of functionality that is not applicable.

[0229] In one option, each AI / ML model capability configuration is associated with at least one identifier (e.g., a configuration ID (e.g., a reporting configuration ID)). The UE includes the identifier when indicating an AI / ML model capability that is not applicable. For example, the UE may have been configured with the following reporting configurations or predicted reporting configurations:

[0230] ■For a)

[0231] Forecast report configuration ID=1

[0232] Forecast report configuration ID=2

[0233] ■For b)

[0234] Forecast report configuration ID=3

[0235] Forecast report configuration ID=4

[0236] ■For c)

[0237] Forecast report configuration ID=5

[0238] Forecast report configuration ID=6

[0239] o Since c) does not apply, the UE includes the following indication in the Complete message:

[0240] Forecast report configuration ID=5->Not applicable;

[0241] Forecast report configuration ID=6->Not applicable;

[0242] o Alternatively, multiple indications regarding the applicable AI / ML model functionality that the UE has configured or is being configured with. The absence of these indications indicates that the AI / ML model functionality is not applicable.

[0243] -3) Reason value or other applicability information

[0244] The oUE may indicate one or more reason values ​​associated with the AI / ML model functionality reported as not applicable.

[0245] Examples of reason values ​​could be:

[0246] ■ Inapplicable network locations, e.g., the UE is connected to a cell where its AI / ML model functionality does not provide output and / or provides output with low or unsuitable accuracy and / or high error and / or uncertainty;

[0247] ■ AI / ML model outputs are not applicable. For example, the model cannot provide a confidence measure for its predictions. The network needs to make use of these predictions. For example, the confidence of the predictions can be used to set the number of beams to be transmitted.

[0248] The AI / ML model is too old. The network may determine that a model older than a certain time is no longer valid (for example, due to network changes).

[0249] ■ Inapplicable UE configuration, for example, the UE has been configured to report frequency layers for which the AI / ML model function was not trained;

[0250] ■ Inappropriate network configuration, for example, the UE connects to the network with a configuration for which the AI / ML model functionality was not trained;

[0251] ■ Inapplicable UE mobility criteria, e.g. the UE has trained the model using measurements and it is a slow UE (via classification based on measurements of the UE's internal sensors and / or via speedStateReselectionPars broadcast by the serving cell), while the UE's current mobility class is high speed.

[0252] ■ Inappropriate CSI-RS configurations, for example, the network is providing CSI-RS resources for channel estimation in a configuration that the AI / ML model function was not trained for (e.g., periodic or aperiodic configuration);

[0253] ■ Inappropriate computing resource availability, e.g., the prediction is not accurate enough due to limited computing resources at the UE (which may be affected by the UE configuration and traffic pattern).

[0254] ■ Inapplicable UE mobility criteria, e.g. the UE has trained the model using measurements and it is a slow UE (via classification based on measurements of the UE's internal sensors and / or via speedStateReselectionPars broadcast by the serving cell), while the UE's current mobility class is high speed.

[0255] - Suggestions / recommendations or reconfigurations;

[0256] The oUE may indicate one or more recommendations or reconfiguration suggestions to make the AI / ML model functionally applicable. For example:

[0257] ■New frequency layer;

[0258] ■ New reference signal configuration for reporting and / or performing inference and / or prediction. For example, the network may have configured the UE to perform time-domain prediction of CSI-RS measurements, but for that cell, the AI / ML model functionality does not apply to CSI-RS but does apply to SSBs, so the UE needs to indicate this to the network. The UE can, for example, suggest a new CSI-RS periodicity.

[0259] ■New community?

[0260] ■New MIMO configuration? CSI-MeasConfig?

[0261] ■What else?

[0262] ■Configured with data collection enabled (e.g., in BM use case, NW scans all beams). This enables the model to be updated and adapted to the current scenario.

[0263] ■ An indication of when the AI / ML model functionality is expected to be applicable (e.g., this may depend on a temporary shortage of computing resources or a newly received configuration). In another example, the UE may indicate that the AI / ML model functionality is applicable when the UE enters a low mobility state.

[0264] In the above-described method, the applicability information (or indication that an AI / ML model function is not applicable) reported by the UE indicates that at least one AI / ML model function configured or being configured by the UE is not applicable. For example, upon determining that all AI / ML models associated with a function for which the UE is trained are not applicable, the UE reports the applicability information (as disclosed in the previous embodiments) for the associated AI / ML model function in a message (such as the message disclosed in the previous embodiments), for example, by indicating that the AI / ML model function is not applicable. In response, the target network node receiving the applicability information may further request information regarding the AI / ML model function, such as in a UE Information Request message. In response, the UE sends further applicability information (e.g., in a UE Information Response message), such as a reason value for the inapplicability and / or one or more recommended configurations. In the applicability information message, the UE may include one or more recommendations or suggestions for reconfiguration to make at least one of the multiple AI / ML models associated with the AI / ML model function applicable.

[0265] 5 Other Solutions

[0266] In one set of embodiments, the network configures the UE to report applicability information of at least one associated AI / ML model. When the UE is not configured to report applicability information of at least one associated AI / ML model, and the UE determines that a given AI / ML model that the UE has configured and / or is configuring is not applicable, the UE performs any one or more of the following actions:

[0267] When the UE is in the connected state, the UE initiates the RRC re-establishment procedure;

[0268] When the UE is in the connected state, the UE declares that the reconfiguration has failed and triggers the RRC re-establishment procedure.

[0269] When the UE is transitioning from the IDLE or INACTIVE state to the CONNECTED state, if security is activated, the UE declares the RRC re-establishment procedure, otherwise, the UE returns to the IDLE state and indicates the failure to the upper layer.

[0270] 6Other descriptions

[0271] Figure 6 An example of a communication system 600 is shown in accordance with some embodiments.

[0272] In the example, communication system 600 includes a telecommunications network 602, which includes an access network 604 (such as a radio access network (RAN)); and a core network 606, which includes one or more core network nodes 608. Access network 604 includes one or more access network nodes, such as network nodes 610A and 610B (one or more of which may be generally referred to as network nodes 610), or any other similar Third Generation Partnership Project (3GPP) access nodes or non-3GPP access points (APs). Furthermore, those skilled in the art will appreciate that network nodes are not necessarily limited to implementations that are provided by a single vendor and integrate radio and baseband portions. Thus, it will be understood that network nodes include disaggregated implementations or portions thereof. For example, in some embodiments, telecommunications network 602 includes one or more open RAN (ORAN) network nodes. An ORAN network node is a node in the telecommunications network 602 that supports ORAN specifications (e.g., specifications promulgated by the O-RAN Alliance or any similar organization) and can operate independently or in conjunction with other nodes to implement one or more functions of any node in the telecommunications network 602 (including one or more network nodes 610 and / or core network nodes 608).

[0273] Examples of ORAN network nodes include an open radio unit (O-RU), an open distributed unit (O-DU), an open central unit (O-CU), a RAN intelligent controller (near real-time or non-real-time) including an O-CU control plane (O-CU-CP) or an O-CU user plane (O-CU-UP), a hosted software or software plug-in (e.g., a near real-time control application (e.g., xApp) or a non-real-time control application (e.g., rApp)), or any combination thereof (the adjective "open" indicates support for ORAN specifications). A network node may support the specifications by, for example, supporting interfaces defined by the ORAN specifications (e.g., A1, F1, W1, E1, E2, X2, Xn interfaces), an open fronthaul user plane interface, or an open fronthaul management plane interface. Furthermore, an ORAN access node may be a logical node within a physical node. Furthermore, an ORAN network node may be implemented in a virtualized environment (described further below) in which one or more network functions are virtualized. For example, the virtualized environment may include an open cloud (O-Cloud) computing platform orchestrated by a service management and orchestration framework via an O-2 interface defined by the O-RAN Alliance or equivalent technology. Network node 610 facilitates direct or indirect connection of user equipment (UE), such as connecting UE 612A, 612B, 612C, and 612D (one or more of which may be generally referred to as UE 612) to core network 606 via one or more wireless connections.

[0274] Example wireless communications over wireless connections include sending and / or receiving wireless signals using electromagnetic waves, radio waves, infrared waves, and / or other types of signals suitable for sending information without the use of wiring, cables, or other material conductors. Additionally, in various embodiments, the communication system 600 may include any number of wired or wireless networks, network nodes, UEs, and / or any other components or systems that may facilitate or participate in the communication of data and / or signals (whether via a wired or wireless connection). The communication system 600 may include and / or interface with any type of communication, telecommunication, data, cellular, radio network, and / or other similar type of system.

[0275] The UE 612 may be any of a variety of communication devices, including wireless devices that are arranged, configured, and / or operable to wirelessly communicate with the network node 610 and other communication devices. Similarly, the network node 610 is arranged, capable, configured, and / or operable to communicate directly or indirectly with the UE 612 and / or with other network nodes or devices in the telecommunications network 602 to enable and / or provide network access (e.g., wireless network access) and / or to perform other functions (e.g., management) in the telecommunications network 602.

[0276] In the depicted example, core network 606 connects network node 610 to one or more hosts, such as host 616. These connections can be direct or indirect via one or more intermediate networks or devices. In other examples, the network node can be directly coupled to the host. Core network 606 includes one or more core network nodes (e.g., core network node 608) comprised of hardware and software components. The features of these components can be substantially similar to those described with respect to the UE, network nodes, and / or hosts, such that the descriptions are generally applicable to the corresponding components of core network node 608. Example core network nodes include functionality of one or more of the following: a mobile switching center (MSC), a mobility management entity (MME), a home subscriber server (HSS), an access and mobility management function (AMF), a session management function (SMF), an authentication server function (AUSF), a subscription identifier dehiding function (SIDF), a unified data management (UDM), a security edge protection proxy (SEPP), a network exposure function (NEF), and / or a user plane function (UPF).

[0277] The host 616 may be owned or controlled by a service provider other than the operator or provider of the access network 604 and / or the telecommunications network 602, and may be operated by or on behalf of the service provider. The host 616 may host a variety of applications to provide one or more services. Examples of such applications include real-time and pre-recorded audio / video content, data collection services (e.g., retrieving and compiling data about various environmental conditions detected by multiple UEs), analytics functions, social media, functions for controlling or otherwise interacting with remote devices, functions for an alarm and monitoring center, or any other such functions performed by a server.

[0278] As a whole, Figure 6The communication system 600 enables connections between UEs, network nodes, and hosts. In this sense, the communication system 600 can be configured to operate according to predefined rules or procedures such as a specific standard, including but not limited to: Global System for Mobile Communications (GSM); Universal Mobile Telecommunications System (UMTS); Long Term Evolution (LTE) and / or other suitable second generation (2G), third generation (3G), fourth generation (4G), or fifth generation (5G) standards, or any applicable future generation standards (e.g., sixth generation (6G)); Wireless Local Area Network (WLAN) standards such as the Institute of Electrical and Electronics Engineers (IEEE) 802.11 standard (WiFi); and / or any other suitable wireless communication standards such as Worldwide Interoperability for Microwave Access (WiMax), Bluetooth, Z-Wave, Near Field Communication (NFC) ZigBee, LiFi, and / or any low power wide area network (LPWAN) standards such as LoRa and Sigfox.

[0279] In some examples, telecommunication network 602 is a cellular network that implements 3GPP standardized features. Thus, telecommunication network 602 can support network slicing to provide different logical networks to different devices connected to telecommunication network 602. For example, telecommunication network 602 can provide ultra-reliable low-latency communication (URLLC) services to some UEs while providing enhanced mobile broadband (eMBB) services to other UEs, and / or provide massive machine type communication (mMTC) / massive Internet of Things (IoT) services to yet other UEs.

[0280] In some examples, the UE 612 is configured to send and / or receive information without direct human interaction. For example, the UE can be designed to send information to the access network 604 according to a predetermined schedule when triggered by an internal or external event or in response to a request from the access network 604. In addition, the UE can be configured to operate in a single radio access technology (RAT) mode or a multi-RAT mode or a multi-standard mode. For example, the UE can operate using any one or a combination of WiFi, New Radio (NR), and LTE, i.e., be configured for multi-radio dual connectivity (MR-DC), such as Evolved UMTS Terrestrial RAN (E-UTRAN) NR Dual Connectivity (EN-DC).

[0281] In an example, a hub 614 communicates with the access network 604 to facilitate indirect communication between one or more UEs (e.g., UE 612C and / or 612D) and a network node (e.g., network node 610B). In some examples, the hub 614 can be a controller, a router, a content source and analyzer, or any other communication device described herein with respect to a UE. For example, the hub 614 can be a broadband router that enables the UE to access the core network 606. As another example, the hub 614 can be a controller that sends commands or instructions to one or more actuators of the UE. The commands or instructions can be received from the UE, the network node 610, or through executable code, scripts, processes, or other instructions in the hub 614. As another example, the hub 614 can be a data collector that acts as a temporary storage device for UE data and, in some embodiments, can perform analysis or other processing on the data. As another example, the hub 614 can be a content source. For example, for a UE that is a virtual reality (VR) headset, display, speaker, or other media delivery device, the center 614 can retrieve VR assets, video, audio, or other media or data related to sensory information via a network node, and then the center 614 provides it directly to the UE after performing local processing and / or adding additional local content. In yet another example, the center 614 acts as a proxy server or orchestrator for the UE, especially if one or more of the UEs are low-energy IoT devices.

[0282] Hub 614 can have a continuous / persistent or intermittent connection to network node 610B. Hub 614 can also enable different communication schemes and / or scheduling between hub 614 and UEs (e.g., UE 612C and / or UE 612D), as well as between hub 614 and core network 606. In other examples, hub 614 connects to core network 606 and / or one or more UEs via a wired connection. Furthermore, hub 614 can be configured to connect to a machine-to-machine (M2M) service provider via access network 604 and / or to another UE via a direct connection. In some scenarios, a UE can establish a wireless connection with network node 610 while still connecting through hub 614 via a wired or wireless connection. In some embodiments, hub 614 can be a dedicated hub, i.e., a hub whose primary function is to route communications from network node 610B to UEs / from UEs to network node 110B. In other embodiments, the center 614 may be a non-dedicated center, ie, a device operable to route communications between the UE and the network node 610B but additionally operable as a communications origin and / or endpoint for certain data channels.

[0283] Figure 7A UE 700 according to some embodiments is shown. As used herein, a UE refers to a device capable of, configured, arranged, and / or operable to wirelessly communicate with a network node and / or other UEs. Examples of UEs include, but are not limited to, smartphones, mobile phones, cellular phones, Voice over Internet Protocol (VoIP) phones, wireless local loop phones, desktop computers, personal digital assistants (PDAs), wireless cameras, gaming consoles or devices, music storage devices, playback devices, wearable terminal devices, wireless endpoints, mobile stations, tablet computers, laptop computers, laptop embedded devices (LEEs), laptop mounted devices (LMEs), smart devices, wireless customer premises equipment (CPEs), vehicles, vehicle-mounted or vehicle-embedded / integrated wireless devices, and the like. Other examples include any UE identified by 3GPP, including narrowband Internet of Things (NB-IoT) UEs, machine type communication (MTC) UEs, and / or enhanced MTC (eMTC) UEs.

[0284] A UE may, for example, support device-to-device (D2D) communication by implementing 3GPP standards for sidelink communication, dedicated short-range communication (DSRC), vehicle-to-vehicle (V2V), vehicle-to-infrastructure (V2I), or vehicle-to-everything (V2X). In other examples, a UE may not necessarily have a user in the sense of a human user who owns and / or operates the associated device. Alternatively, a UE may represent a device that is intended for sale to or operated by a human user but may not, or may not initially, be associated with a specific human user (e.g., a smart sprinkler controller). Alternatively, a UE may represent a device that is not intended for sale to or operated by an end user but may be associated with or operated for the benefit of a user (e.g., a smart power meter).

[0285] UE 700 includes a processing circuit 702 operatively coupled to an input / output interface 706, a power supply 708, a memory 710, a communication interface 712, and / or any other components or any combination thereof via a bus 704. Some UEs may utilize Figure 7 All or a subset of the components shown. The level of integration between components may vary depending on the UE. In addition, some UEs may include multiple instances of components, such as multiple processors, memories, transceivers, transmitters, receivers, etc.

[0286] Processing circuitry 702 is configured to process instructions and data and may be configured to implement any sequential state machine operable to execute instructions stored as a machine-readable computer program in memory 710. Processing circuitry 702 may be implemented as: one or more hardware-implemented state machines (e.g., implemented in discrete logic, a field-programmable gate array (FPGA), an application-specific integrated circuit (ASIC), etc.); programmable logic along with appropriate firmware; one or more stored computer programs, general-purpose processors (such as microprocessors or digital signal processors (DSPs)) along with appropriate software; or any combination of the foregoing. For example, processing circuitry 702 may include multiple central processing units (CPUs).

[0287] In an example, the input / output interface 706 may be configured to provide one or more interfaces to an input device, an output device, or one or more input and / or output devices. Examples of output devices include a speaker, a sound card, a video card, a display, a monitor, a printer, an actuator, a transmitter, a smart card, another output device, or any combination thereof. An input device may allow a user to capture information into the UE 700. Examples of input devices include a touch-sensitive or presence-sensitive display, a camera (e.g., a digital camera, a digital video camera, a webcam, etc.), a microphone, a sensor, a mouse, a trackball, a directional pad, a trackpad, a scroll wheel, a smart card, etc. A presence-sensitive display may include a capacitive or resistive touch sensor to sense input from the user. The sensor may be, for example, an accelerometer, a gyroscope, a tilt sensor, a force sensor, a magnetometer, an optical sensor, a proximity sensor, a biometric sensor, etc., or any combination thereof. The output device may use the same type of interface port as the input device. For example, a Universal Serial Bus (USB) port may be used to provide both input and output devices.

[0288] In some embodiments, power supply 708 is configured as a battery or battery pack. Other types of power sources may be used, such as an external power source (e.g., a power outlet), a photovoltaic device, or a battery. Power supply 708 may also include power circuitry for delivering power from power supply 708 itself and / or an external power source to various components of UE 700 via an input circuit or an interface such as a power cable. The delivered power may be used, for example, to charge power supply 708. The power circuitry may perform any formatting, conversion, or other modifications on the power from power supply 708 to make the power suitable for the respective components of UE 700 to which it is supplied.

[0289] The memory 710 may be or be configured to include a memory such as a random access memory (RAM), a read-only memory (ROM), a programmable ROM (PROM), an erasable PROM (EPROM), an electrically EPROM (EEPROM), a magnetic disk, an optical disk, a hard disk, a removable tape, a flash drive, etc. In one example, the memory 710 includes one or more application programs 714, such as an operating system, a web browser application, a widget, a gadget engine, or other applications, and corresponding data 716. The memory 710 may store any one of a variety of operating systems or a combination of operating systems used by the UE 700.

[0290] The memory 710 may be configured to include multiple physical drive units, such as a redundant array of independent disks (RAID), a flash memory, a USB flash drive, an external hard drive, a thumb drive, a pen drive, a key drive, a high-density digital versatile disc (HD-DVD) optical drive, an internal hard drive, a Blu-ray disc drive, a holographic digital data storage (HDDS) optical drive, an external mini dual in-line memory module (DIMM), a synchronous dynamic random access memory (SDRAM), an external micro-DIMM SDRAM, a smart card memory (e.g., a tamper-resistant module in the form of a universal integrated circuit card (UICC) including one or more subscriber identity modules (SIMs) such as a universal SIM (USIM) and / or an Internet Protocol Multimedia Services identity module (ISIM), other memories, or any combination thereof. The UICC may be, for example, an embedded UICC (eUICC), an integrated UICC (iUICC), or a removable UICC commonly referred to as a "SIM card." The memory 710 may allow the UE to 700 accesses instructions, applications, etc. stored on a temporary or non-temporary storage medium to download data or upload data. An article of manufacture (such as an article of manufacture utilizing a communication system) may be tangibly embodied as or in a memory 710, which may be or include a device-readable storage medium.

[0291] The processing circuit 702 can be configured to communicate with an access network or other network using a communication interface 712. The communication interface 712 may include one or more communication subsystems and may include an antenna 722 or be communicatively coupled to the antenna 722. The communication interface 712 may include one or more transceivers for communication (such as by communicating with one or more remote transceivers of another device capable of wireless communication (e.g., another UE or a network node in the access network)). Each transceiver may include a transmitter 718 and / or a receiver 720 suitable for providing network communication (e.g., optical, electrical, frequency allocation, etc.). In addition, the transmitter 718 and the receiver 720 may be coupled to one or more antennas (e.g., antenna 722) and may share circuit components, software, or firmware, or alternatively be implemented separately.

[0292] In the illustrated embodiment, the communication functionality of the communication interface 712 may include cellular communication, WiFi communication, LPWAN communication, data communication, voice communication, multimedia communication, short-range communication such as Bluetooth, NFC, location-based communication (e.g., using a global positioning system (GPS) to determine location), another type of communication functionality, or any combination thereof. Communication may be implemented according to one or more communication protocols and / or standards, such as IEEE 802.11, Code Division Multiple Access (CDMA), Wideband CDMA (WCDMA), GSM, LTE, New Radio (NR), UMTS, WiMax, Ethernet, Transmission Control Protocol / Internet Protocol (TCP / IP), Synchronous Optical Network (SONET), Asynchronous Transfer Mode (ATM), Quick User Datagram Protocol Internet Connection (QUIC), Hypertext Transfer Protocol (HTTP), etc.

[0293] Regardless of the type of sensor, the UE can provide an output of the data captured by its sensor via its communication interface 712 via a wireless connection to a network node. The data captured by the UE's sensor can be transmitted via another UE via a wireless connection to a network node. The output can be periodic (e.g., once every 15 minutes if it reports the sensed temperature), random (e.g., to balance the load of reports from several sensors), in response to a trigger event (e.g., sending an alert when humidity is detected), in response to a request (e.g., a user-initiated request), or a continuous stream (e.g., a live video feed of a patient).

[0294] As another example, a UE includes an actuator, motor, or switch associated with a communication interface configured to receive wireless input from a network node via a wireless connection. In response to the received wireless input, the state of the actuator, motor, or switch can change. For example, the UE can include a motor that adjusts the control surfaces or rotors of a drone in flight based on the received input, or adjusts a robotic arm performing a medical procedure based on the received input.

[0295] When the UE is in the form of an IoT device, the UE may be a device used in one or more application areas including, but not limited to, urban wearable technology, expanded industrial applications, and healthcare. Non-limiting examples of such IoT devices are or are embedded in: an internet-connected refrigerator or freezer, a television, internet-connected lighting, an electric meter, a robotic vacuum cleaner, a voice-controlled smart speaker, a home security camera, a motion detector, a thermostat, a smoke detector, a door and window sensor, a flood / humidity sensor, an electronic door lock, an internet-connected doorbell, an air conditioning system (such as a heat pump), an autonomous vehicle, a monitoring system, a weather monitoring device, a vehicle parking monitoring device, an electric vehicle charging station, a smartwatch, a fitness tracker, a head-mounted display for augmented reality (AR) or VR, a wearable device for tactile or sensory enhancement, a sprinkler, an animal tracking or item tracking device, a sensor for monitoring plants or animals, an industrial robot, an unmanned aerial vehicle (UAV), and any kind of medical device (such as a heart rate monitor or a teleoperated surgical robot). In addition to the above, Figure 7 In addition to the other components depicted in the illustrated UE 700 , a UE in the form of an IoT device may also include circuitry and / or software depending on the intended application of the IoT device.

[0296] As yet another specific example, in an IoT scenario, a UE may represent a machine or other device that performs monitoring and / or measurements and transmits the results of such monitoring and / or measurements to another UE and / or a network node. In this case, the UE may be an M2M device, which in the 3GPP context may be referred to as an MTC device. As a specific example, the UE may implement the 3GPP NB-IoT standard. In other scenarios, a UE may represent a vehicle (such as a car, bus, truck, ship, or airplane) or other device capable of monitoring and / or reporting its operating status or other functions associated with its operation.

[0297] In practice, any number of UEs can be used together for a single use case. For example, the first UE could be a drone or integrated into a drone, and provide the drone's speed information (obtained via a speed sensor) to a second UE, which is a remote controller operating the drone. When the user makes changes from the remote controller, the first UE can adjust the drone's throttle (e.g., by controlling an actuator) to increase or decrease the drone's speed. The first and / or second UEs can also include more than one of the aforementioned functionalities. For example, a UE could include both a sensor and an actuator, and handle data communications for both the speed sensor and the actuator.

[0298] Figure 8 A network node 800 according to some embodiments is shown. As used herein, a network node refers to a device capable of, configured, arranged, and / or operable to communicate directly or indirectly with a UE and / or with other network nodes or devices in a telecommunications network. Examples of network nodes include, but are not limited to, APs (e.g., radio APs), base stations (BSs) (e.g., radio BSs, Node Bs, evolved Node Bs (eNBs), NR Node Bs (gNBs)), and O-RAN nodes or components of O-RAN nodes (e.g., O-RUs, O-DUs, O-CUs).

[0299] Base stations can be categorized based on the amount of coverage they provide (or, in other words, their transmit power level), and thus, depending on the amount of coverage provided, they can be referred to as femto, pico, micro, or macro base stations. A base station can be a relay node or a relay donor node that controls a relay. A network node can also include one or more (or all) parts of a distributed radio base station, such as a centralized digital unit, a distributed unit (e.g., in an O-RAN access node), and / or a remote radio unit (RRU), sometimes referred to as a remote radio head (RRH). These RRUs may or may not be integrated with antennas as antenna-integrated radios. Parts of a distributed radio base station can also be referred to as nodes in a distributed antenna system (DAS).

[0300] Other examples of network nodes include a multi-transmission point (multi-TRP) 5G access node, a multi-standard radio (MSR) device (such as an MSR BS), a network controller (such as a radio network controller (RNC) or a BS controller (BSC)), a base transceiver station (BTS), a transmission point, a transmission node, a multi-cell / multicast coordination entity (MCE), an operation and maintenance (O&M) node, an operation support system (OSS) node, a self-organizing network (SON) node, a positioning node (e.g., an evolved serving mobile location center (E-SMLC)) and / or a minimization of drive tests (MDT).

[0301] Network node 800 includes processing circuitry 802, memory 804, a communication interface 806, and a power supply 808. Network node 800 may be comprised of multiple physically separate components (e.g., a Node B component and an RNC component, a BTS component and a BSC component, etc.), each of which may have its own corresponding components. In some scenarios where network node 800 includes multiple separate components (e.g., a BTS component and a BSC component), one or more separate components may be shared across multiple network nodes. For example, a single RNC may control multiple NodeBs. In such scenarios, each unique NodeB and RNC pair may, in some cases, be considered a separate network node. In some embodiments, network node 800 may be configured to support multiple RATs. In such embodiments, some components may be replicated (e.g., separate memory 804 may exist for different RATs), and some components may be reused (e.g., the same antenna 810 may be shared by different RATs). The network node 800 may also include multiple sets of the various components shown for different wireless technologies (e.g., GSM, WCDMA, LTE, NR, WiFi, Zigbee, Z-wave, Long Range Wide Area Network (LoRaWAN), Radio Frequency Identification (RFID), or Bluetooth wireless technology) integrated into the network node 800. These wireless technologies may be integrated into the same or different chips or chipsets and other components within the network node 800.

[0302] The processing circuitry 802 may include a combination of one or more of the following: a microprocessor, a controller, a microcontroller, a CPU, a DSP, an ASIC, an FPGA, or any other suitable computing device, resource, or combination of hardware, software, and / or encoded logic operable to provide network node 800 functionality, alone or in combination with other network node 800 components (e.g., memory 804).

[0303] In some embodiments, processing circuitry 802 comprises a system on a chip (SOC). In some embodiments, processing circuitry 802 comprises one or more of radio frequency (RF) transceiver circuitry 812 and baseband processing circuitry 814. In some embodiments, RF transceiver circuitry 812 and baseband processing circuitry 814 may be on separate chips (or chipsets), boards, or units (e.g., a radio unit and a digital unit). In alternative embodiments, some or all of RF transceiver circuitry 812 and baseband processing circuitry 814 may be on the same chip, chipset, board, or unit.

[0304] Memory 804 may include any form of volatile or non-volatile computer-readable memory, including, but not limited to, permanent storage devices, solid-state memory, remotely mounted memory, magnetic media, optical media, RAM, ROM, mass storage media (e.g., a hard drive), removable storage media (e.g., a flash drive, compact disc (CD), or digital video disc (DVD)), and / or any other volatile memory or non-volatile, non-transitory device-readable and / or computer-executable memory device that stores information, data, and / or instructions that can be used by processing circuitry 802. Memory 804 may store any suitable instructions, data, or information, including computer programs, software, applications including one or more of logic, rules, code, tables, and / or other instructions that can be executed by processing circuitry 802 and used by network node 800. Memory 804 may be used to store any computations performed by processing circuitry 802 and / or any data received via communication interface 806. In some embodiments, processing circuitry 802 and memory 804 are integrated.

[0305] Communication interface 806 is used for wired or wireless communication of signaling and / or data between network nodes, access networks, and / or UEs. As shown, communication interface 806 includes port / terminal 816 for sending and receiving data to and from the network, for example, via a wired connection. Communication interface 806 also includes radio front-end circuitry 818, which can be coupled to antenna 810, or in some embodiments, to a portion of antenna 810. Radio front-end circuitry 818 includes filter 820 and amplifier 822. Radio front-end circuitry 818 can be connected to antenna 810 and processing circuitry 802. Radio front-end circuitry 818 can be configured to condition signals transmitted between antenna 810 and processing circuitry 802. Radio front-end circuitry 818 can receive digital data to be transmitted to other network nodes or UEs via a wireless connection. Radio front-end circuitry 818 can use a combination of filter 820 and / or amplifier 822 to convert the digital data into a radio signal with appropriate channel and bandwidth parameters. The radio signal can then be transmitted via antenna 810. Similarly, when data is received, antenna 810 may collect the radio signal, which may then be converted to digital data by radio front-end circuitry 818. The digital data may be passed to processing circuitry 802. In other embodiments, communication interface 806 may include different components and / or different combinations of components.

[0306] In certain alternative embodiments, network node 800 does not include separate radio front-end circuitry 818; instead, processing circuitry 802 includes the radio front-end circuitry and is connected to antenna 810. Similarly, in some embodiments, all or some of RF transceiver circuitry 812 is part of communication interface 806. In yet another embodiment, communication interface 806 includes one or more ports or terminals 816, radio front-end circuitry 818, and RF transceiver circuitry 812 as part of a radio unit (not shown), and communication interface 806 communicates with baseband processing circuitry 814, which is part of a digital unit (not shown).

[0307] Antenna 810 may include one or more antennas or antenna arrays configured to transmit and / or receive wireless signals. Antenna 810 may be coupled to radio front-end circuitry 818 and may be any type of antenna capable of wirelessly transmitting and receiving data and / or signals. In some embodiments, antenna 810 is separate from network node 800 and may be connected to network node 800 via an interface or port.

[0308] The antenna 810, the communication interface 806, and / or the processing circuit 802 may be configured to perform any receiving operations and / or certain obtaining operations described herein performed by the network node 800. Any information, data, and / or signals may be received from a UE, another network node, and / or any other network device. Similarly, the antenna 810, the communication interface 806, and / or the processing circuit 802 may be configured to perform any transmitting operations described herein performed by the network node 800. Any information, data, and / or signals may be transmitted to a UE, another network node, and / or any other network device.

[0309] Power supply 808 provides power to the various components of network node 800 in a form suitable for the various components (e.g., at the voltage and current levels required by each corresponding component). Power supply 808 may also include or be coupled to power management circuitry to supply power to the components of network node 800 for performing the functions described herein. For example, network node 800 may be connected to an external power source (e.g., an electrical grid or an electrical outlet) via an input circuit or interface (such as a cable), whereby the external power source supplies power to the power circuitry of power supply 808. As another example, power supply 808 may include a power source in the form of a battery or battery pack that is connected to or integrated into the power circuitry. The battery may provide backup power if the external power source fails.

[0310] Embodiments of network node 800 may include more than Figure 8Components in addition to the components shown are used to provide certain aspects of the functionality of the network node (including any functionality described herein and / or any functionality required to support the subject matter described herein). For example, network node 800 may include a user interface device to allow information to be input into network node 800 and to allow information to be output from network node 800. This may allow a user to perform diagnostic, maintenance, repair, and other management functions with respect to network node 800.

[0311] Figure 9 is a block diagram of a host 900 according to various aspects described herein, which may be Figure 6 As used herein, host 900 may be or include various combinations of hardware and / or software, including standalone servers, blade servers, cloud-enabled servers, distributed servers, virtual machines, containers, or processing resources in a server farm. Host 900 may provide one or more services to one or more UEs.

[0312] Host 900 includes processing circuitry 902 operatively coupled to input / output interface 906, network interface 908, power supply 910, and memory 912 via bus 904. Other components may be included in other embodiments. The features of these components may be substantially similar to those described with respect to previous figures (such as Figure 7 and Figure 8 ) so that its description is generally applicable to corresponding components of the host 900.

[0313] Memory 912 may include one or more computer programs, including data 916, which may include user data, such as data generated by a UE for host 900, or data generated by host 900 for a UE, and one or more host applications 914. Embodiments of host 900 may utilize only a subset or all of the components shown. Host applications 914 may be implemented in a container-based architecture and may provide support for video codecs (e.g., Versatile Video Coding (VVC), High Efficiency Video Coding (HEVC), Advanced Video Coding (AVC), Moving Picture Experts Group (MPEG), VP9) and audio codecs (e.g., Free Lossless Audio Codec (FLAC), Advanced Audio Coding (AAC), MPEG, G.711), including transcoding for a variety of different classes, types, or implementations of UEs (e.g., mobile phones, desktop computers, wearable display systems, and head-up display systems). Host applications 914 may also provide user authentication and permission checks and may periodically report health, routing, and content availability to a central node (e.g., a device in the core network or at the edge of the core network). Thus, the host 900 can select and / or instruct the UE on different hosts for over-the-top (OTT) services. The host application 914 can support various protocols, such as HTTP Live Streaming (HLS), Real-Time Messaging Protocol (RTMP), Real-Time Streaming Protocol (RTSP), Dynamic Adaptive Streaming over HTTP (DASH or MPEG-DASH), etc.

[0314] Figure 10 is a block diagram illustrating a virtualized environment 1000 in which functionality implemented by some embodiments can be virtualized. In this context, virtualization means creating a virtual version of an apparatus or device that may include virtualized hardware platforms, storage devices, and network resources. As used herein, virtualization may apply to any device or component thereof described herein and refers to embodiments in which at least a portion of the functionality is implemented as one or more virtual components. Some or all of the functionality described herein may be implemented as virtual components executed by one or more virtual machines (VMs) implemented in one or more virtual environments 1000 hosted by one or more hardware nodes (such as hardware computing devices operating as network nodes, UEs, core network nodes, or hosts). Furthermore, in embodiments where a virtual node does not require radio connectivity (e.g., a core network node or host), the node may be fully virtualized. In some embodiments, virtualized environment 1000 includes components defined by the O-RAN Alliance, such as an open cloud environment orchestrated by a service management and orchestration framework via the O-2 interface.

[0315] An application 1002 (which may alternatively be referred to as a software instance, a virtual application, a network function, a virtual node, a virtual network function, etc.) runs in the virtualized environment 1000 to implement some features, functions, and / or benefits of some embodiments disclosed herein.

[0316] The hardware 1004 includes processing circuitry, memory storing software and / or instructions that can be executed by the hardware processing circuitry, and / or other hardware devices described herein (such as network interfaces, input / output interfaces, etc.). The software can be executed by the processing circuitry to instantiate one or more virtualization layers 1006 (also referred to as hypervisors or virtual machine monitors (VMMs)), provide VMs 1008a and 1008b (one or more of which can be generally referred to as VMs 1008), and / or perform any of the functions, features, and / or benefits described in connection with some embodiments described herein. The virtualization layer 1006 can present a virtual operating platform to the VMs 1008 that appears to be network hardware.

[0317] VM 1008 includes virtual processing, virtual memory, virtual networks or interfaces, and virtual storage, and can be run by a corresponding virtualization layer 1006. Different embodiments of instances of virtual device 1002 can be implemented on one or more VMs 1008, and these implementations can be made in different ways. In some contexts, virtualization of hardware is referred to as network function virtualization (NFV). NFV can be used to unify many types of network equipment into industry-standard high-capacity server hardware, physical switches, and physical storage, which can be located in data centers and customer premises equipment.

[0318] In the context of NFV, VMs 1008 can be software implementations of physical machines whose operating programs behave as if they were executed on a physical, non-virtualized machine. Each VM 1008 and the portion of hardware 1004 that executes that VM (whether dedicated to that VM and / or shared with other VMs 1008) form a separate virtual network element. Still in the context of NFV, a virtual network function is responsible for handling specific network functions operating in one or more VMs 1008 on hardware 1004 and corresponding to applications 1002.

[0319] Hardware 1004 can be implemented in a standalone network node with general or specialized components. Hardware 1004 may implement some functionality via virtualization. Alternatively, hardware 1004 may be part of a larger hardware cluster (e.g., in a data center or CPE), where many hardware nodes work together and are managed by management and coordination 1010, which oversees, among other things, the lifecycle management of applications 1002. In some embodiments, hardware 1004 is coupled to one or more radio units, each of which includes one or more transmitters and one or more receivers that may be coupled to one or more antennas. Radio units can communicate directly with other hardware nodes via one or more appropriate network interfaces and can be used in conjunction with virtual components to provide radio capabilities to virtual nodes (e.g., RAN or base stations). In some embodiments, some signaling can be provided through the use of a control system 1012, which can alternatively be used for communication between hardware nodes and radio units.

[0320] Figure 11 A communication diagram is shown in which a host 1102 communicates with a UE 1106 via a network node 1104 over a partially wireless connection according to some embodiments. Figure 11 Describe the UE discussed in the previous paragraph (such as Figure 6 UE 612A and / or Figure 7 UE 700), network nodes (such as Figure 6 network node 610A and / or Figure 8 network nodes 800) and hosts (such as Figure 6 Host 616 and / or Figure 9 An example implementation of a host 900 according to various embodiments.

[0321] Similar to host 900, embodiments of host 1102 include hardware, such as a communication interface, processing circuitry, and memory. Host 1102 also includes software that is stored in or accessible by host 1102 and executed by the processing circuitry. The software includes a host application that is operable to provide services to a remote user, such as UE 1006, connected via an OTT connection 1150 extending between UE 1106 and host 1102. In providing services to the remote user, the host application can provide user data sent using OTT connection 1150.

[0322] The network node 1104 includes hardware that enables it to communicate with the host 1102 and the UE 1106. The connection 1160 can be a direct connection or through a core network (such as Figure 6The core network 606 of the network and / or one or more other intermediate networks (such as one or more public, private or managed networks). For example, the intermediate network can be a backbone network or the Internet.

[0323] UE 1106 includes hardware and software, the software being stored in or accessible by UE 1106 and executable by the UE's processing circuitry. This software includes a client application (such as a web browser or operator-specific "app") operable to provide services to a human or non-human user via UE 1106, with support from host 1102. Within host 1102, an executing host application can communicate with an executing client application via an OTT connection 1150, which terminates between UE 1106 and host 1102. When providing services to a user, the UE's client application can receive request data from the host application and provide user data in response to the request data. The OTT connection 1150 can transmit both the request data and the user data. The UE's client application can interact with the user to generate user data that is provided to the host application via the OTT connection 1150.

[0324] The OTT connection 1150 may extend via a connection 1160 between the host 1102 and the network node 1104 and via a wireless connection 1170 between the network node 1104 and the UE 1106 to provide connectivity between the host 1102 and the UE 1106. The connection 1160 and the wireless connection 1170 over which the OTT connection 1150 may be provided have been drawn abstractly to illustrate communication between the host 1102 and the UE 1106 via the network node 1104, without explicitly involving any intermediary devices and the precise routing of messages via those devices.

[0325] As an example of transmitting data via OTT connection 1150, in step 1108, host 1102 provides user data, which may be performed by executing a host application. In some embodiments, the user data is associated with a specific human user who interacts with UE 1106. In other embodiments, the user data is associated with UE 1106, which shares data with host 1102 without explicit human interaction. In step 1110, host 1102 initiates a transmission to UE 1106, carrying the user data. Host 1102 may initiate the transmission in response to a request sent by UE 1106. The request may be caused by human interaction with UE 1106 or by operation of a client application executing on UE 1106. In accordance with the teachings of the embodiments described throughout this disclosure, the transmission may be delivered via network node 1104. Thus, in step 1112, in accordance with the teachings of the embodiments described throughout this disclosure, network node 1104 sends the user data carried in the transmission initiated by host 1102 to UE 1106. In step 1114, UE 1106 receives the user data carried in the transmission, which may be performed by a client application executing on UE 1106 that is associated with a host application executed by host 1102.

[0326] In some examples, UE 1106 executes a client application that provides user data to host 1102. The user data may be provided as a reaction or response to data received from host 1102. Thus, in step 1116, UE 1106 may provide the user data, which may be performed by executing the client application. When providing the user data, the client application may also consider user input received from the user via an input / output interface of UE 1106. Regardless of the specific manner in which the user data is provided, in step 1118, UE 1106 initiates a transmission of the user data to host 1102 via network node 1104. In step 1120, network node 1104 receives the user data from UE 1106 and initiates transmission of the received user data to host 1102 in accordance with the teachings of the embodiments described throughout this disclosure. In step 1122, host 1102 receives the user data carried in the transmission initiated by UE 1106.

[0327] One or more of the various embodiments improve the performance of OTT services provided to the UE 1106 using the OTT connection 1150 in which the wireless connection 1170 forms the final part.

[0328] In an example scenario, the host 1102 may collect and analyze plant status information. As another example, the host 1102 may process audio and video data that may have been retrieved from the UE for use in creating a map. As another example, the host 1102 may collect and analyze real-time data to help control vehicle congestion (e.g., controlling traffic lights). As another example, the host 1102 may store surveillance video uploaded by the UE. As another example, the host 1102 may store or control access to media content such as video, audio, VR, or AR, which may be broadcast, multicast, or unicast to the UE. As other examples, the host 1102 may be used for energy pricing, remote control of non-time-critical electrical loads to balance power generation demand, positioning services, presentation services (e.g., compiling charts based on data collected from remote devices, etc.), or any other function that collects, retrieves, stores, analyzes, and / or transmits data.

[0329] In some examples, a measurement process may be provided for monitoring data rate, latency, and other factors targeted for improvement in one or more embodiments. Optional network functionality may also be present for reconfiguring the OTT connection 1150 between the host 1102 and the UE 1106 in response to changes in measurement results. The measurement process and / or network functionality for reconfiguring the OTT connection 1150 may be implemented in software and hardware on the host 1102 and / or the UE 1106. In some embodiments, sensors (not shown) may be deployed in or associated with other devices through which the OTT connection 1150 passes. The sensors may participate in the measurement process by providing values ​​for the monitored quantities listed above or other physical quantities from which the software can calculate or estimate the monitored quantities. Reconfiguration of the OTT connection 1150 may include message formats, retransmission settings, preferred routing, and the like; reconfiguration does not require direct changes to the operation of the network node 1104. Such processes and functionality may be known and practiced in the art. In some embodiments, the measurements may involve proprietary UE signaling that facilitates the host 1102's measurement of throughput, propagation time, latency, and the like. Measurements can be made by software using the OTT connection 1150 to send messages (particularly empty or "dummy" messages) while monitoring propagation times, errors, etc.

[0330] While the computing devices (e.g., UEs, network nodes, hosts) described herein may include combinations of the hardware components shown, other embodiments may include computing devices with different combinations of components. It should be understood that these computing devices may include any suitable combination of hardware and / or software necessary to perform the tasks, features, functions, and methods disclosed herein. The determinations, calculations, acquisitions, or similar operations described herein may be performed by processing circuitry, which may process information by, for example, converting the acquired information into other information, comparing the acquired or converted information with information stored in the network node, and / or performing one or more actions based on the acquired or converted information, and making determinations based on the results of the processing. Furthermore, while components are depicted as a single block within a larger block or nested within multiple blocks, in reality, a computing device may include multiple different physical components that make up the single illustrated component, and functionality may be divided between separate components. For example, a communication interface may be configured to include any of the components described herein, and / or functionality of a component may be divided between the processing circuitry and the communication interface. In another example, non-computationally intensive functionality of any such component may be implemented in software or firmware, while computationally intensive functionality may be implemented in hardware.

[0331] In some embodiments, some or all of the functionality described herein may be provided by a processing circuit that executes instructions stored in a memory, which in some embodiments may be a computer program product in the form of a non-transitory computer-readable storage medium. In alternative embodiments, some or all of the functionality may be provided by the processing circuit 620, for example, in a hardwired manner, without executing instructions stored on a separate or discrete device-readable storage medium. In any of these specific embodiments, the processing circuit may be configured to perform the described functionality, regardless of whether instructions stored on a non-transitory computer-readable storage medium are executed. The benefits provided by such functionality are not limited to separate processing circuits or to other components of the computing device, but are enjoyed by the computing device as a whole and / or generally by end users and wireless networks.

[0332] Example embodiments of the present disclosure are as follows:

[0333] Group A Examples

[0334] Embodiment 1: A method performed by a user equipment (UE) for reporting applicability information of at least one artificial intelligence (AI) or machine learning (ML) model associated with a function, the method comprising: sending (212; 304; 408; 528) applicability information of at least one AI or ML model associated with a function that has been configured and / or is being configured of the UE to a network node.

[0335] Embodiment 2: The method according to embodiment 1, wherein the applicability information of at least one AI or ML model associated with the function includes any one or more of the following items: an indication that at least one AI or ML model associated with the function is not applicable; an indication that at least one AI or ML model associated with the function is not applicable; a reason value; a suggestion, recommendation, or reconfiguration on how to make the at least one AI or ML model applicable.

[0336] Embodiment 3: The method according to embodiment 1 or 2, wherein sending (212; 408; 528) the applicability information includes sending (212; 408; 528) a completion message, wherein the completion message includes applicability information of at least one AI or ML model associated with the functions that have been configured and / or are being configured of the UE.

[0337] Embodiment 4: The method according to embodiment 3, wherein the completion message is any one or more of the following items: an RRC recovery completion message, an RRC establishment completion message, and an RRC reconfiguration completion message.

[0338] Embodiment 5: The method according to embodiment 3 or 4 further includes: sending (200; 400) a request message to a network node; receiving (208; 404) a response message from the network node in response to the request message, wherein sending (212; 408) a completion message including the applicability information includes sending (212; 408) the completion message after receiving the response message.

[0339] Embodiment 6: The method according to embodiment 5, wherein the request message is an RRC recovery request, the response message is an RRC recovery message, and the completion message is an RRC recovery complete message.

[0340] Embodiment 7: The method according to embodiment 5, wherein the request message is an RRC setup request, the response message is an RRC setup message, and the completion message is an RRC setup complete message.

[0341] Embodiment 8: The method according to any one of embodiments 5 to 7, wherein the response message includes configuration information of a function associated with the at least one AI or ML model, or information for activating a function associated with the at least one AI or ML model.

[0342] Embodiment 9: The method according to embodiment 8, wherein sending (212; 408) the completion message includes sending (212; 408) the completion message in response to configuration information of a function associated with the at least one AI or ML model, or information for activating a function associated with the at least one AI or ML model.

[0343] Embodiment 10: The method of embodiment 3 or 4 further comprising: receiving (524) a reconfiguration message from the network node, wherein sending (528) a completion message including the applicability information comprises sending (528) the completion message after receiving the response message.

[0344] Embodiment 11: The method according to embodiment 10, wherein the reconfiguration message is an RRC reconfiguration message, and the completion message is an RRC reconfiguration complete message.

[0345] Embodiment 12: The method according to embodiment 10 or 11, wherein the reconfiguration message includes configuration information of a function associated with the at least one AI or ML model, or information for activating a function associated with the at least one AI or ML model.

[0346] Embodiment 13: The method of embodiment 12, wherein sending (528) the completion message comprises sending (528) the completion message in response to configuration information of a function associated with the at least one AI or ML model, or information for activating a function associated with the at least one AI or ML model.

[0347] Embodiment 14: The method according to embodiment 1 or 2, wherein sending the applicability information includes sending the applicability information in a request message or through a request message.

[0348] Embodiment 15: The method according to embodiment 14, wherein the request message is in an RRC recovery request.

[0349] Embodiment 16: The method according to the preceding embodiment further includes: providing user data; and forwarding the user data to the host via transmission to the network node.

[0350] Group B Examples

[0351] Embodiment 17: A method performed by a network node for obtaining applicability information of at least one artificial intelligence (AI) or machine learning (ML) model associated with a function, the method comprising: receiving (212; 304; 408; 528) applicability information of at least one AI or ML model associated with a function that has been configured and / or is being configured of the UE from a user equipment (UE).

[0352] Embodiment 18: The method according to embodiment 17, wherein the applicability information of at least one AI or ML model associated with the function includes any one or more of the following items: an indication that at least one AI or ML model associated with the function is not applicable; an indication that at least one AI or ML model associated with the function is not applicable; a reason value; a suggestion, recommendation, or reconfiguration on how to make the at least one AI or ML model applicable.

[0353] Embodiment 19: The method according to embodiment 17 or 18, wherein receiving (212; 408; 528) the applicability information includes receiving (212; 408; 528) a completion message, wherein the completion message includes applicability information of at least one AI or ML model associated with the functions that have been configured and / or are being configured of the UE.

[0354] Embodiment 20: The method according to embodiment 19, wherein the completion message is any one or more of the following items: an RRC recovery completion message, an RRC establishment completion message, and an RRC reconfiguration completion message.

[0355] Embodiment 21: The method according to embodiment 19 or 20 further includes: receiving (200; 400) a request message from a UE; and sending (208; 404) a response message to the UE in response to the request message, wherein receiving (212; 408) a completion message including applicability information includes receiving (212; 408) the completion message after sending the response message.

[0356] Embodiment 22: The method according to embodiment 21, wherein the request message is an RRC recovery request, the response message is an RRC recovery message, and the completion message is an RRC recovery complete message.

[0357] Embodiment 23: The method of embodiment 21, wherein the request message is an RRC setup request, the response message is an RRC setup message, and the completion message is an RRC setup complete message.

[0358] Embodiment 24: The method according to any one of Embodiments 21 to 23, wherein the response message includes configuration information of a function associated with the at least one AI or ML model, or information for activating a function associated with the at least one AI or ML model.

[0359] Embodiment 25: The method of embodiment 24, wherein receiving (212; 408) the completion message comprises receiving (212; 408) the completion message in response to configuration information of a function associated with the at least one AI or ML model, or information for activating a function associated with the at least one AI or ML model.

[0360] Embodiment 26: The method of embodiment 19 or 20 further comprising: sending (524) a reconfiguration message to the UE, wherein receiving (528) a completion message including the applicability information comprises receiving (528) the completion message after sending the response message.

[0361] Embodiment 27: The method of embodiment 26, wherein the reconfiguration message is an RRC reconfiguration message, and the completion message is an RRC reconfiguration complete message.

[0362] Embodiment 28: The method according to embodiment 26 or 27, wherein the reconfiguration message includes configuration information of a function associated with the at least one AI or ML model, or information for activating a function associated with the at least one AI or ML model.

[0363] Embodiment 29: The method of embodiment 28, wherein receiving (528) the completion message comprises receiving (528) the completion message in response to configuration information of a function associated with the at least one AI or ML model, or information for activating a function associated with the at least one AI or ML model.

[0364] Embodiment 30: The method according to embodiment 17 or 18, wherein sending the applicability information includes sending the applicability information in a request message or through a request message.

[0365] Example 31: The method according to Example 30, wherein the request message is in an RRC recovery request.

[0366] Embodiment 32: The method according to any one of embodiments 17 to 31 further includes performing one or more actions based on the applicability information.

[0367] Embodiment 33: The method according to any of the preceding embodiments further includes: obtaining user data; and forwarding the user data to a host or user device.

[0368] Group C Examples

[0369] Embodiment 34: A user device comprises: a processing circuit configured to perform any of the steps described in any of the embodiments in Group A; and a power supply circuit configured to supply power to the processing circuit.

[0370] Embodiment 35: A network node, comprising: a processing circuit configured to perform any of the steps described in any of the embodiments in Group B; and a power supply circuit configured to supply power to the processing circuit.

[0371] Embodiment 36: A user equipment (UE) comprising: an antenna configured to send and receive wireless signals; a radio front-end circuit connected to the antenna and the processing circuit and configured to condition the signal transmitted between the antenna and the processing circuit; a processing circuit configured to perform any steps described in any of the embodiments in Group A; an input interface connected to the processing circuit and configured to allow information to be input into the UE for processing by the processing circuit; an output interface connected to the processing circuit and configured to output information from the UE that has been processed by the processing circuit; and a battery connected to the processing circuit and configured to supply power to the UE.

[0372] Embodiment 37: A host configured to operate in a communication system to provide an over-the-top (OTT) service, the host comprising: a processing circuit configured to provide user data; and a network interface configured to initiate sending the user data to a network node in a cellular network for sending to a user equipment (UE), the network node having a communication interface and a processing circuit, the processing circuit of the network node being configured to perform any operation according to any embodiment of Group B to send the user data from the host to the UE.

[0373] Embodiment 38: A host according to the preceding embodiment, wherein: the processing circuit of the host is configured to execute a host application that provides user data; and the UE includes a processing circuit that is configured to execute a client application associated with the host application to receive a transmission of the user data from the host.

[0374] Embodiment 39: A method implemented in a host, the host being configured to operate in a communication system, the communication system also including a network node and a user equipment (UE), the method comprising: providing user data to the UE; and initiating a transmission to the UE via a cellular network including the network node, the transmission carrying the user data, wherein the network node performs any steps described in any of the embodiments in Group B to send the user data from the host to the UE.

[0375] Embodiment 40: The method according to the above embodiment further includes: sending user data provided by the host to the UE at the network node.

[0376] Embodiment 41: A method according to any one of the preceding two embodiments, wherein user data is provided at the host by executing a host application, which interacts with a client application executed on the UE, and the client application is associated with the host application.

[0377] Embodiment 42: A communication system configured to provide an over-the-top (OTT) service, the communication system comprising a host, the host comprising: a processing circuit configured to provide user data associated with the over-the-top service to a user equipment (UE); and a network interface configured to initiate sending the user data to a cellular network node for sending to the UE, the network node having a communication interface and a processing circuit, the processing circuit of the network node being configured to perform any operation described in any of the embodiments in Group B to send the user data from the host to the UE.

[0378] Embodiment 43: The communication system according to the aforementioned embodiment further includes: a network node; and / or a UE.

[0379] Embodiment 44: A host configured to operate in a communication system to provide an over-the-top (OTT) service, the host comprising: a processing circuit configured to initiate reception of user data; and a network interface configured to receive user data from a network node in a cellular network, the network node having a communication interface and a processing circuit, the processing circuit of the network node being configured to perform any operation described in any of the embodiments in Group B to receive user data from a user equipment (UE) of the host.

[0380] Embodiment 45: A host according to the two aforementioned embodiments, wherein: the processing circuit of the host is configured to execute a host application that receives user data, and the host application is configured to interact with a client application executed on the UE, which client application is associated with the host application.

[0381] Embodiment 46: The host according to any one of the preceding two embodiments, wherein initiating reception of user data includes requesting user data.

[0382] Embodiment 47: A method implemented by a host, the host being configured to operate in a communication system further comprising a network node and a user equipment (UE), the method comprising: at the host, initiating reception of user data from the UE, the user data originating from a transmission already received by the network node from the UE, wherein the network node performs any of the steps described in any of the embodiments in Group B to receive user data from the UE for the host.

[0383] Embodiment 48: The method according to the above embodiment further includes: at the network node, sending the received user data to the host.

[0384] Embodiment 49: A host configured to operate in a communication system to provide an over-the-top (OTT) service, the host comprising: a processing circuit configured to provide user data; and a network interface configured to initiate sending the user data to a cellular network for sending to a user equipment (UE), wherein the UE comprises a communication interface and a processing circuit, the communication interface and processing circuit of the UE being configured to perform any operation described in any embodiment of Group A to receive user data from the host.

[0385] Embodiment 50: The host according to the preceding embodiment, wherein the cellular network further comprises a network node configured to communicate with the UE to send user data from the host to the UE.

[0386] Embodiment 51: A host according to the two aforementioned embodiments, wherein: the processing circuit of the host is configured to execute a host application to provide user data, and the host application is configured to interact with a client application executed on the UE, which client application is associated with the host application.

[0387] Embodiment 52: A method implemented by a host operating in a communication system, the communication system also including a network node and a user equipment (UE), the method comprising: providing user data to the UE; and initiating a transmission to the UE via a cellular network including the network node, the transmission carrying the user data, wherein the network node performs any steps described in any embodiment of Group A to receive the user data from the host.

[0388] Embodiment 53: The method according to the aforementioned embodiment further includes: executing, at the host, a host application associated with the client application executed on the UE to receive user data from the host application.

[0389] Embodiment 54: The method according to the aforementioned embodiment further includes: at the host, sending input data to the client application executed on the UE, the input data being provided by executing the host application, wherein the user data is provided by the client application in response to the input data from the host application.

[0390] Embodiment 55: A host configured to operate in a communication system to provide an over-the-top (OTT) service, the host comprising: a processing circuit configured to provide user data; and a network interface configured to initiate sending the user data to a cellular network for sending to a user equipment (UE), wherein the UE comprises a communication interface and a processing circuit, and the communication interface and processing circuit of the UE are configured to perform any steps described in any embodiment of Group A to send user data to the host.

[0391] Embodiment 56: The host according to the preceding embodiment, wherein the cellular network further comprises a network node configured to communicate with the UE to send user data from the UE to the host.

[0392] Embodiment 57: A host according to the two aforementioned embodiments, wherein: the processing circuit of the host is configured to execute a host application to provide user data, and the host application is configured to interact with a client application executed on the UE, which client application is associated with the host application.

[0393] Embodiment 58: A method implemented by a host, the host being configured to operate in a communication system further comprising a network node and a user equipment (UE), the method comprising: receiving, at the host, user data sent by the UE to the host via the network node, wherein the UE performs any steps described in any embodiment of Group A to send the user data to the host.

[0394] Embodiment 59: The method according to the aforementioned embodiment further includes: executing, at the host, a host application associated with the client application executed on the UE to receive user data from the UE.

[0395] Embodiment 60: The method according to the aforementioned two embodiments further includes: at the host, sending input data to the client application executed on the UE, the input data being provided by executing the host application, wherein the user data is provided by the client application in response to the input data from the host application.

[0396] Those skilled in the art will recognize improvements and modifications to the embodiments of the present disclosure, and all such improvements and modifications are considered to fall within the scope of the concepts disclosed herein.

Claims

1. A method, performed by a user equipment (UE), for reporting applicability information of at least one artificial intelligence (AI) or machine learning (ML) model associated with a function, the method comprising: Send to the network node (212; 304; 408; 528) Applicability information of at least one AI or ML model associated with a function that has been configured and / or is being configured on the UE.

2. The method according to claim 1, wherein The applicability information of the at least one AI or ML model associated with the function includes any one or more of the following: an indication that the at least one AI or ML model associated with the functionality is not applicable; an indication that functionality associated with the at least one AI or ML model is not applicable; Reason value; A suggestion, recommendation, or reconfiguration of how to adapt the at least one AI or ML model.

3. The method according to claim 1 or 2, wherein: Sending (212; 408; 528) the applicability information includes sending (212; 408; 528) a completion message, the completion message including the applicability information of the at least one AI or ML model associated with the configured and / or being configured functionality of the UE.

4. The method according to claim 3, wherein: The completion message is a radio resource control RRC resumption complete message, an RRC establishment complete message, or an RRC reconfiguration complete message.

5. The method according to claim 3 or 4, further comprising: Sending a (200; 400) request message to the network node; In response to the request message, receiving from the network node (208; 404) response message; Wherein, sending (212; 408) the completion message including the applicability information comprises: sending (212; 408) the completion message after receiving the response message.

6. The method according to claim 5, wherein: The request message is an RRC recovery request, the response message is an RRC recovery message, and the completion message is an RRC recovery complete message.

7. The method according to claim 5, wherein: The request message is an RRC setup request, the response message is an RRC setup message, and the completion message is an RRC setup complete message.

8. The method according to any one of claims 5 to 7, wherein The response message includes configuration information of a function associated with the at least one AI or ML model, or information for activating a function associated with the at least one AI or ML model.

9. The method according to claim 8, wherein Sending (212; 408) the completion message includes sending (212; 408) the completion message in response to configuration information of a function associated with the at least one AI or ML model, or information for activating a function associated with the at least one AI or ML model.

10. The method according to claim 3 or 4, further comprising: receiving (524) a reconfiguration message from the network node; Wherein sending (528) the completion message including the applicability information includes sending (528) the completion message after receiving the response message.

11. The method according to claim 10, wherein: The reconfiguration message is an RRC reconfiguration message, and the completion message is an RRC reconfiguration complete message.

12. The method according to claim 10 or 11, wherein: The reconfiguration message includes configuration information of a function associated with the at least one AI or ML model, or information for activating a function associated with the at least one AI or ML model.

13. The method according to claim 12, wherein: Sending (528) the completion message includes sending (528) the completion message in response to configuration information of a function associated with the at least one AI or ML model, or information for activating a function associated with the at least one AI or ML model.

14. The method according to claim 1 or 2, wherein: Sending the applicability information includes sending the applicability information in a request message or through a request message.

15. The method according to claim 14, wherein The request message is in the RRC recovery request.

16. A user equipment (UE) (612; 700) for reporting applicability information of at least one artificial intelligence (AI) or machine learning (ML) model associated with a function, the UE being adapted to: Applicability information of at least one AI or ML model associated with a configured and / or being configured function of the UE is sent (212; 304; 408; 528) to a network node.

17. The UE according to claim 16, further adapted to perform the method according to any one of claims 2 to 15.

18. A user equipment (UE) (612; 700) for reporting applicability information of at least one artificial intelligence (AI) or machine learning (ML) model associated with a function, the UE (612; 700) comprising: a communication interface (712) comprising a transmitter (718) and a receiver (720); as well as A processing circuit (702) associated with the communication interface (712), the processing circuit (702) being configured to cause the UE (612; 700) to send (212; 304; 408; 528) to a network node applicability information of at least one AI or ML model associated with a function that has been configured and / or is being configured of the UE.

19. The UE according to claim 18, wherein: The processing circuit (702) is further configured to cause the UE (612; 700) to perform the method according to any one of claims 2 to 15.

20. A method performed by a network node for obtaining suitability information of at least one artificial intelligence (AI) or machine learning (ML) model associated with a function, the method comprising: Received from user equipment (UE) (212; 304； 408; 528) Applicability information of at least one AI or ML model associated with a function that has been configured and / or is being configured on the UE.

21. The method according to claim 20, wherein The applicability information of the at least one AI or ML model associated with the function includes any one or more of the following: an indication that the at least one AI or ML model associated with the functionality is not applicable; an indication that functionality associated with the at least one AI or ML model is not applicable; Reason value; A suggestion, recommendation, or reconfiguration of how to adapt the at least one AI or ML model.

22. The method according to claim 20 or 21, wherein Receiving (212; 408; 528) the applicability information includes receiving (212; 408; 528) a completion message, the completion message including the applicability information of the at least one AI or ML model associated with the configured and / or being configured functionality of the UE.

23. The method according to claim 22, wherein The completion message is any one or more of the following: an RRC recovery complete message, an RRC setup complete message, and an RRC reconfiguration complete message.

24. The method according to claim 22 or 23, further comprising: receiving (200; 400) a request message from the UE; In response to the request message, sending (208; 404) response message; Wherein, receiving (212; 408) a completion message including the applicability information comprises: receiving (212; 408) the completion message after sending the response message.

25. The method according to claim 24, wherein The request message is an RRC recovery request, the response message is an RRC recovery message, and the completion message is an RRC recovery complete message.

26. The method according to claim 24, wherein The request message is an RRC setup request, the response message is an RRC setup message, and the completion message is an RRC setup complete message.

27. The method according to any one of claims 24 to 26, wherein The response message includes configuration information of a function associated with the at least one AI or ML model, or information for activating a function associated with the at least one AI or ML model.

28. The method according to claim 27, wherein Receiving (212; 408) the completion message includes receiving (212; 408) the completion message in response to configuration information of a function associated with the at least one AI or ML model, or information for activating a function associated with the at least one AI or ML model.

29. The method according to claim 22 or 23, further comprising: sending (524) a reconfiguration message to the UE; Wherein, receiving (528) a completion message including the applicability information includes receiving (528) the completion message after sending the response message.

30. The method according to claim 29, wherein The reconfiguration message is an RRC reconfiguration message, and the completion message is an RRC reconfiguration complete message.

31. The method according to claim 29 or 30, wherein The reconfiguration message includes configuration information of a function associated with the at least one AI or ML model, or information for activating a function associated with the at least one AI or ML model.

32. The method according to claim 31, wherein Receiving (528) the completion message includes receiving (528) the completion message in response to configuration information of a function associated with the at least one AI or ML model, or information for activating a function associated with the at least one AI or ML model.

33. The method according to claim 20 or 21, wherein Sending the applicability information includes sending the applicability information in a request message or through a request message.

34. The method according to claim 33, wherein The request message is in the RRC recovery request.

35. The method of any one of claims 20 to 34, further comprising performing one or more actions based on the suitability information.

36. A network node (610; 800) for obtaining suitability information of at least one artificial intelligence (AI) or machine learning (ML) model associated with a function, the network node (610; 800) being adapted to: Applicability information of at least one AI or ML model associated with a configured and / or being configured functionality of a user equipment (UE) is received (212; 304; 408; 528).

37. The network node (610; 800) according to claim 36, further adapted to perform the method according to any one of claims 21 to 35.

38. A network node (610; 800) for obtaining suitability information of at least one artificial intelligence (AI) or machine learning (ML) model associated with a function, the network node (610; 800) comprising: Communication interface (806); as well as A processing circuit (802) associated with a communication interface (806), the processing circuit (802) being configured to cause the network node (610; 800) to receive (212; 304; 408; 528) from a user equipment (UE) suitability information of at least one AI or ML model associated with a configured and / or being configured function of the UE.

39. The network node (610; 800) according to claim 38, further adapted to perform the method according to any one of claims 21 to 35.

Images