Method, apparatus and system for enhancing model applicability

The method enhances model applicability in wireless communication systems by dynamically switching between primary and lightweight models based on device and network context, addressing inefficiencies and performance degradation in existing systems.

WO2026234211A1PCT designated stage Publication Date: 2026-11-12CONTINENTAL AUTOMOTIVE TECHNOLOGIES GMBH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/EP2026/063347
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2025-05-08
Filing Date
2026-05-08
Publication Date
2026-11-12

AI Technical Summary

Technical Problem

Existing wireless communication systems face challenges in efficiently managing and adapting artificial intelligence and machine learning models due to device resource constraints, leading to performance degradation and inefficiencies in model applicability.

Method used

A method and apparatus for enhancing model applicability by allowing wireless devices to dynamically switch between primary and lightweight models based on device and network context, using configuration and control information from network nodes to optimize resource usage and maintain functionality.

Benefits of technology

This approach improves flexibility and reduces performance degradation by enabling adaptive model selection and efficient switching between models, ensuring appropriate model usage based on current device conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure EP2026063347_12112026_PF_FP_ABST
    Figure EP2026063347_12112026_PF_FP_ABST
Patent Text Reader

Abstract

A method, performed by a wireless device for enhancing model applicability in a wireless communication system is provided. A wireless device receives configuration information of one or more models from a network node, wherein the one or more models comprise a first model. The wireless device determines a first applicability status of the first model. The wireless device transmits an indication of the first applicability status to the network node. The wireless device receives, from the network node, control information related to a second model. The wireless device switches between the first model and the second model based on the control information.
Need to check novelty before this filing date? Find Prior Art

Description

2025026441Method, Apparatus and System for Enhancing Model ApplicabilityTECHNICAL FIELD

[0001] The disclosure relates to wireless communication technology, and more specifically to methods and apparatus for enhancing model applicability in a wireless communication system involving wireless devices and network nodes.BACKGROUND

[0002] Wireless communication systems are widely deployed to provide various telecommunication services such as telephony, video, data, messaging, broadcasts, or other similar types of services. An example of a communication network would be a 3rd Generation Partnership Project (3GPP) 5G (fifth generation) New Radio (NR) standardbased telecommunications network.

[0003] Wireless communication systems typically comprise interconnected network entities and user devices that exchange information over radio interfaces. These systems enable the transmission and reception of data, voice, and multimedia content between users and network infrastructure.SUMMARY

[0004] According to a first aspect of the present disclosure, a method, performed by a wireless device for enhancing model applicability in a wireless communication system is provided. The method comprises receiving configuration information of one or more models from a network node, wherein the one or more models comprise a first model. The method further comprises determining a first applicability status of the first model. The method further comprises transmitting an indication of the first applicability status to the network node. The method further comprises receiving, from the network node, control information related to a second model. The method further comprises switching between the first model and the second model based on the control information.

[0005] The method provides improved flexibility in managing artificial intelligence and machine learning models in wireless communication systems. For example, the method may allow dynamic adaptation of model usage based on device and / or network context, and / or may reduce performance degradation by enabling efficient switching between models according to control information received from the network node.Internal2025026442

[0006] The wireless communication system may refer to a network comprising at least one network node and at least one wireless device, wherein the network node transmits configuration information and control information to the wireless device, and the wireless device transmits indications of model applicability status to the network node. The wireless communication system may be based on standards such as 3GPP 5G NR, LTE, Wi-Fi, or any other protocol supporting radio access.

[0007] The wireless device may refer to a wireless communication unit capable of communicating with a network node in a wireless communication system, such as a mobile phone, tablet, laptop, loT device, or any other device equipped with a radio transceiver. For example, the wireless device may refer to user equipment (UE).

[0008] The network node may refer to any type of network node. It can transmit and receive radio signals, manage communication resources, and facilitate the exchange of data and control information with the wireless device (e.g. user equipment). The network node (e.g. base station) may serve as an access point for user devices, enabling mobile connectivity and supporting various network services. For example, the network node refers to a wireless communication unit such as a base station.

[0009] Enhancing model applicability may mean increasing the ability of the wireless device (e.g. the user equipment) to utilize an appropriate model for a given context, such as device resource availability or network policy, thereby maintaining or improving the performance of AI / ML-enabled features or functionalities. Model applicability may refer to the suitability or feasibility of executing a particular model on the wireless device under current conditions, such as available compute resources, memory, or battery level.

[0010] The configuration information of one or more models may refer to data transmitted from the network node to the wireless device, specifying parameters, metadata, or operational details for at least one artificial intelligence (Al) or machine learning (ML) model. The configuration information may comprise model identifiers, inference settings, hardware requirements, or any other information necessary for model execution. For example, the wireless device receives configuration information of both first (primary or applicable) model and second (light-weight) model.

[0011] The first model may refer to an artificial intelligence or machine learning model configured for use in the wireless device, and may be selected based on device context and / or network instructions. The first applicability status may refer to a determination by the wireless device regarding whether the first model can be executed based on current device resources, such as compute resources, memory, or battery, or based on network context or model metadata. The indication of the first applicability status may be aInternal2025026443message or signal transmitted to the network node, and may comprise a single-bit value or any other format suitable for reporting applicability.

[0012] The control information related to a second model may refer to data transmitted from the network node to the wireless device, specifying activation, switching, or configuration instructions for an alternative model. In an example, the control information may instruct the wireless device to switch from the first model to the second model. In another example, the control information may instruct the wireless device to switch back from the second model to the first model. The second model may refer to a model that is different from the first model, and may be a light-weight model or any other model suitable for execution under constrained device conditions.

[0013] Switching between the first model and the second model based on the control information may mean that the wireless device activates or deactivates the execution of the respective models in response to the received control information. The switching may be performed automatically based on the control information and may depend on the current context of the wireless device, such as improved device resources or a change in applicability status.

[0014] In some embodiments, the one or more models may include a primary model and a light-weight model. Preferably, the first model may be the primary model, and the second model may be the light-weight model.

[0015] The wireless device could switch between the primary model and the lightweight model, which improves: flexibility in switching between models of differing complexity or resource requirements, maintenance of functionality under device constraints, and optimization of performance and resource usage.

[0016] In some embodiments, the primary model may be a designated model which is an AI / ML model designated as primary by the network node, or a full model which is an AI / ML model intended for use under normal device conditions and providing a full feature and / or functionality set. The light-weight model may be an alternative AI / ML model intended for use when the primary model is inapplicable due to device resource constraints and may provide at least part of the feature and / or functionality set of the primary model.

[0017] A primary model (also referred to as an applicable model) may refer to an artificial intelligence or machine learning model configured to provide comprehensive functionality and optimal performance within a wireless communication system. This model is intended for use when device resources, such as processing power, memory, and battery, are sufficient to support its operation.Internal2025026444

[0018] A light-weight model may refer to an alternative artificial intelligence or machine learning model designed with reduced resource requirements. The light-weight model is intended for use when device resources are constrained, enabling continued operation of essential features with lower performance or accuracy compared to the primary (applicable) model. Both models are associated with the same functionality, allowing dynamic adaptation to varying device conditions.

[0019] In some embodiments, the indication of the first applicability status may include a single-bit value indicating inapplicability. The indication of the first applicability status may be transmitted via uplink control information, UCI, or uplink MAC control element, UL MAC CE. The control information may comprise switching information related to the second model. The indication of the first applicability status may include a cause of the first applicability status

[0020] The indication of the first applicability status may include a single-bit value indicating inapplicability, which could enable minimal signaling overhead for reporting model inapplicability. The indication of the first applicability status may be transmitted via uplink control information or uplink MAC control element, allowing use of established uplink signaling mechanisms for efficient communication between the wireless device (or the UE) and the network node (or the base station). The control information may include switching information related to the second model, thereby supporting rapid and context-aware switching between models.

[0021] In some embodiments, the control information may include configuration information of the second model.

[0022] For example, the wireless device (or the UE) receives configuration information of the first model from the base station. After receiving the indication of the first applicability status, the base station could determine whether to provide the control information. If the indication indicates that the first model is inapplicable in the wireless device, the base station could send the control information including the configuration of the second model to the wireless device. In this way, the wireless device does not need to receive massive configuration information when the first model is applicable, which saves computational resources.

[0023] In some embodiments, the receiving configuration information of one or more models from the network node may include receiving configuration information of the first and second models.

[0024] For example, the wireless device (e.g. the UE) could receive the configuration information of the primary and light-models together, which enables the wireless device toInternal2025026445access multiple model configurations simultaneously. This arrangement may facilitate faster switching between models, reduce latency associated with requesting additional configurations, and improve adaptability to changes in wireless device context or network instructions.

[0025] In some embodiments, the control information may include an activation / switching command related to the second model.

[0026] The control information may include an activation or switching command related to the second model, enabling direct and immediate control over model selection at the wireless device. This arrangement may allow the network node to instruct the wireless device to activate or switch between the first model and the second model without delay. For example, when the first model is inapplicable in the wireless device, the wireless device may receive control information including command of switching from the first model to the second model from the network node. For another example, when the first model is applicable again in the wireless device, the wireless device may receive control information including command of switching back from the second model to the first model from the network node.

[0027] In some embodiments, the transmitting an indication of the first applicability status to the network node may include: transmitting the indication of the first applicability status to the network node when the first model is inapplicable in the wireless device.

[0028] This arrangement could ensure that the signaling of model applicability status is event-driven and occurs only upon detection of inapplicability, rather than being sent periodically or in all cases. Such event-driven approach could reduce unnecessary signaling.

[0029] In some embodiments, the determining a first applicability status of a first model comprises determining the first applicability status based on any of wireless device context, network context, or model metadata. The wireless device context may comprise any of compute resources, memory availability, or battery level. The network context may comprise any of Al model policy, gNB configuration, or slice profile. The model metadata may comprise any of model size, execution time estimate, or hardware dependency.

[0030] The wireless device context may refer to operational parameters of the wireless device, such as the amount of available processing capability, the current state of memory resources, or the remaining battery charge. For example, the wireless device context may refer to the UE context.

[0031] For example, the wireless device may initially operate an applicable (primary) model to provide an AI / ML-enabled feature. If the wireless device context changes — suchInternal2025026446as a reduction in available compute resources, limited memory, or a low battery level — the wireless device may determine that the applicable model can no longer be executed efficiently. Upon receiving this indication, the network node provides control information to the wireless device. The wireless device then switches from the applicable model to the light-weight model, thereby maintaining the essential functionality with reduced resource requirements. When the wireless device context improves, for instance, after charging the battery or freeing up memory, the wireless device may again transmit an indication to the network node and switch back to the applicable model based on updated control information.

[0032] This arrangement could eliminate performance degradation or resource exhaustion by ensuring that the selected model is appropriate for the current device conditions.

[0033] According to a second aspect of the present disclosure, a method, performed by a network node, for enhancing model applicability in a wireless communication system is provided. The method comprises transmitting configuration information for one or more models to a wireless device, wherein the one or more models comprise a first model. The method further comprises receiving, from the wireless device, an indication of a first model applicability status. The method comprises determining, based on the indication, whether to provide control information related to a second model. The method further comprises transmitting the control information to the wireless device to trigger model switching.

[0034] The method provides improved flexibility in adapting model usage in a wireless communication system. For example, the method may allow dynamic switching between different models at the wireless device based on applicability status, and / or may reduce performance degradation by enabling the network node to transmit control information for model switching.

[0035] The network node may refer to a wireless communication unit or any type of network node configured to manage wireless communication with wireless device, such as a base station, a Next generation Node B (gNB), an eNodeB, or any other radio access network node capable of transmitting and receiving configuration and control information. The network node may include a processor, a memory, and a communication interface, or any other component capable of performing the method steps.

[0036] Transmitting configuration information for one or more models to the wireless device may refer to sending data specifying parameters, settings, or operational details for at least one model, such as an AI / ML model, over a wireless interface. The configuration information may comprise model metadata, execution parameters, or activationInternal2025026447commands, and may be sent via radio resource control messages, downlink control information, or downlink MAC control elements.

[0037] The one or more models may refer to a set of models, such as a first model and a second model, where each model is an AI / ML model or any other computational model used for wireless communication features or functionalities. The first model may be a fullfeature model intended for normal device conditions, and the second model may be a light-weight model intended for use when the first model is inapplicable due to device resource constraints. In one example, the first model may not necessarily be the fullfeature model; it could be selected or designated by the network node for a particular use case, device type, or operational scenario (e.g., a model optimized for automotive applications or a model tailored for a specific slice profile).

[0038] Receiving, from the wireless device, an indication of a first model applicability status may refer to the network node obtaining a message or signal from the wireless device that specifies whether the first model is applicable or inapplicable. The indication may include a single-bit value, such as “0” for applicability and “1” for inapplicability, and may be transmitted via uplink control information or uplink MAC control elements.

[0039] Transmitting the control information to the wireless device to trigger model switching may refer to the network node sending configuration information or commands that allow the wireless device to switch between the first model and the second model. The control information may include configuration details, activation or switching commands, switch back commands, and may be sent via downlink control information or downlink MAC control elements. The model switching may be based on the wireless device context and the received control information.

[0040] In some embodiments, the control information may include configuration information of the second model and / or an activation / switching command related to the second model. The control information may be transmitted via downlink control information, DCI, or downlink MAC control element, DL MAC CE. The network node may correspond to a Next generation Node B, gNB, configured to communicate the at least one input signal to the wireless device.

[0041] In some embodiments, the method further comprises receiving, from the wireless device, a subsequent indication, wherein the subsequent indication indicates that the first model is applicable in the wireless device. The method may comprise transmitting a switch back command to the wireless device, wherein the switch back command triggers the wireless device to switch back from the second model to the first model.

[0042] The subsequent indication may be provided when the wireless device context,Internal2025026448such as compute resources, memory, or battery level, has improved to allow use of the first model. The switch back command may be transmitted via downlink control information or a downlink MAC control element. This approach could provide dynamic adaptation of model selection in the wireless device (or the UE).

[0043] According to a third aspect of the present application, a wireless device for enhancing model applicability in a wireless communication system is provided. The wireless device comprises a communication module configured to receive configuration information for one or more models from a network node, wherein the one or more models comprise a first model. The communication module is further configured to transmit an indication of a first model applicability status to the network node. The communication module is also configured to receive, from the network node, control information related to a second model. The wireless device comprises a processing module configured to determine the first applicability status of the first model. The processing module is further configured to switch between the first model and the second model based on the control information.

[0044] The wireless device improves the flexibility and efficiency of deploying artificial intelligence or machine learning models in wireless communication systems. For example, the user equipment may provide adaptive selection and use of models based on device and network context, and / or may reduce performance degradation by switching to a more suitable model when the current model is inapplicable.

[0045] The wireless device may refer to a device configured for wireless communication with a network node, such as a mobile phone, tablet, wireless modem, or any other device capable of operating in a wireless communication system. The wireless device may comprise hardware and / or software components, such as a processor, memory, and radio transceiver, or any other component capable of supporting wireless communication and model processing. For example, the wireless device may refer to user equipment.

[0046] The communication module may refer to an element configured to exchange information with a base station or network node, such as a transceiver, network interface, or a combination of hardware and software components. The communication module may comprise a radio frequency front-end, a baseband processor, and protocol stack software, or any other component capable of transmitting and receiving configuration and control information.

[0047] The configuration information for one or more models may refer to data specifying parameters, structure, or operational details of artificial intelligence or machineInternal2025026449learning models, such as model weights, architecture, version, or resource requirements. The configuration information may be transmitted from the network node to the wireless device using signaling protocols, such as RRC (Radio Resource Control) messages, or any other suitable communication method.

[0048] The first model may refer to an artificial intelligence or machine learning model initially configured for use by the wireless device, which may be a full-featured model or a designated model intended for normal or designated device conditions. The first model may comprise a set of algorithms, parameters, or instructions executable by the processing module.

[0049] The indication of a first model applicability status may refer to a message or signal transmitted from the wireless device to the network node, indicating whether the first model is currently applicable or inapplicable. The indication may comprise a single-bit value, a status code, or any other data format suitable for conveying applicability information.

[0050] The network node may refer to the wireless communication infrastructure, such as a base station, core network, or any other entity capable of receiving applicability status and providing control information.

[0051] The control information related to a second model may refer to data or commands transmitted from the network node to the wireless device, instructing the wireless device to activate, configure, or switch between different AI / ML models. The control information may comprise activation commands, configuration parameters, or switching instructions.

[0052] The second model may refer to an alternative artificial intelligence or machine learning model, such as a light-weight model intended for use when the first model is inapplicable due to device resource constraints. The second model may provide reduced or partial functionality compared to the first model.

[0053] The processing module may refer to an element configured to execute instructions, process data, and manage model applicability and switching, such as a central processing unit, digital signal processor, or any other component capable of performing model evaluation and control.

[0054] The determination of the first applicability status may refer to an evaluation performed by the processing module, based on device context such as compute resources, memory availability, battery level, or other operational parameters. The determination may involve monitoring device status, comparing resource requirements, or executing diagnostic routines.Internal20250264410

[0055] The switching between the first model and the second model based on the control information may refer to the activation, deactivation, or replacement of models in the wireless device, as directed by the received control information. The switching may be performed by loading, unloading, or reconfiguring model parameters in the processing module.

[0056] In an example, the wireless device performs the first model. The processing module determines that the first model (primary / applicable model) is inapplicable in the wireless device. The communication module transmits an indication of the inapplicability to the network node and receives control information of switching to the second model (lightweight model) from the network node. The processing module switches from the first model to the second model. The configuration information of the second model can be transmitted together with that of the first model or as part of the control information.

[0057] In another example, the wireless device performs the second model. The processing module determines that the first model is applicable in the UE. The communication module transmits an indication of the applicability to the network node and receives control information of switching back from the second model to the first model from the network node. The processing module switches from the second model to the first model.

[0058] The communication module and the processing module may operate together, wherein the communication module exchanges information with the network node and the processing module evaluates applicability and manages model switching based on received data. The communication module may provide received configuration and control information to the processing module, and the processing module may instruct the communication module to transmit applicability status.

[0059] According to the fourth aspect of the present disclosure, a network node for enhancing model applicability in a wireless communication system is provided. The network node comprises a communication module configured to transmit configuration information for one or more models to a wireless device, wherein the one or more models comprise a first model. The communication module is further configured to receive, from the wireless device, an indication of a first model applicability status. The communication module is also configured to transmit control information to the wireless device to trigger model switching. The network node further comprises a processing module configured to determine, based on the indication, whether to provide control information related to a second model.

[0060] The network node improves adaptive management of model applicability inInternal20250264411wireless communication systems. For example, the network node may dynamically control the provision of control information for model switching in response to wireless device context and model applicability status, which may reduce performance degradation and allow more efficient configuration of model parameters.

[0061] The network node may refer to a wireless communication unit or any type of network node configured to manage wireless communication with wireless device, such as a Next Generation Node B (gNB) or any other radio access network node capable of transmitting and receiving configuration and control information related to machine learning models. The network node may comprise hardware and / or software components, such as processors, memory, and communication interfaces, or any other component capable of performing the described functions. For example, the network node may refer to a base station.

[0062] The communication module may refer to an element configured to handle the transmission and reception of information between the network node and the wireless device. The communication module may comprise a transceiver, a network interface, or any other component capable of exchanging configuration information, applicability indications, and control information with the wireless device.

[0063] The configuration information for one or more models may refer to data specifying operational parameters, structure, or activation details for machine learning models intended for use by the wireless device. The configuration information may comprise model weights, architecture details, activation commands, or any other data necessary for the wireless device to utilize the models.

[0064] The one or more models may refer to machine learning models, such as artificial intelligence models, that are intended to provide specific features or functionalities in the wireless communication system. The first model may refer to a primary or default model initially configured for the wireless device, while the second model may refer to an alternative model, such as a light-weight model, that may be activated when the first model is inapplicable.

[0065] The wireless device may refer to a wireless communication unit capable of communicating with the network node, such as a mobile phone, tablet, loT device, or any other device with wireless communication capability and support for machine learning model execution. For example, the wireless device may refer to user equipment (UE), and the wireless device context may refer to UE context.

[0066] The indication of a first model applicability status may refer to a message or signal sent from the wireless device to the network node, representing whether the firstInternal20250264412model is currently applicable or inapplicable based on the wireless device context (or the user equipment context). The indication may comprise a single-bit value or any other data format suitable for conveying applicability status.

[0067] The control information to trigger model switching may refer to data or commands transmitted from the network node to the wireless device, instructing the wireless device to activate, deactivate, or switch between machine learning models. The control information may comprise configuration data for the second model, activation commands, or any other information necessary for model switching.

[0068] In one example, the communication module of the network node transmits configuration information of both the first and the second model to the wireless device. After receiving the indication of inapplicability of the first model from the wireless device, the communication module of the network node transmits the control information including activation / switch command to the wireless device. In another example, the communication module at first transmits configuration information of the first model to the wireless device. After receiving the indication of inapplicability of the first model from the wireless device, the communication module of the network node transmits the control information including the configuration information of the second model to the wireless device.

[0069] The processing module may refer to an element within the network node configured to process received indications of model applicability status and determine whether to provide control information related to a second model. The processing module may comprise a processor, logic circuitry, or any other component capable of executing the described decision-making function.

[0070] The determination based on the indication may refer to the processing module evaluating the received applicability status from the wireless device and deciding whether to transmit control information for model switching.

[0071] The communication module and the processing module may be implemented as separate hardware or software components, or as integrated elements within the network node, provided that the described functions are performed. The communication between the modules may be realized via internal data buses, software interfaces, or any other suitable means.

[0072] The transmission of configuration information, reception of applicability status, and transmission of control information may occur over wireless communication channels, such as those defined by 3GPP standards, or any other protocol suitable for exchanging such information between the network node and the wireless device.

[0073] According to a fifth aspect of the present disclosure, a wireless communicationInternal20250264413system for enhancing model applicability is provided. The system includes: at least one wireless device according to the first aspect and any of its embodiments; at least one network node according to the second aspect and any of its embodiments. The at least one wireless device and the at least one network node are capable of being coupled via at least one of wired coupling and / or wireless coupling

[0074] The wireless communication system may have at least one or all the optional features described above.

[0075] The wireless device may refer to a device configured to communicate with the network node, such as a mobile phone, tablet, loT sensor, or vehicle communication unit. The network node may refer to a network node configured to manage wireless communication with wireless device, such as a gNodeB, eNodeB, or access point. For example, the wireless device may refer to a user equipment, while the network node may refer to a base station.

[0076] The coupling may refer to a connection that enables data exchange between the wireless device (or the user equipment) and the network node (or the base station), which may comprise a wired interface, such as Ethernet or fiber optic, or a wireless interface, such as 5G NR, Wi-Fi, or Bluetooth. The system may comprise both types of coupling simultaneously or selectively, depending on network requirements.

[0077] According to a sixth aspect of the present disclosure, a non-transitory computer-readable medium is provided. The non-transitory computer-readable medium stores instructions that, when executed by a processor, cause the processor to perform the method according to any of the first aspect, the second aspect and their embodiment.

[0078] According to a seventh aspect of the present disclosure, a computer program product is provided. The computer program product comprises instructions which, when executed by at least one processor, configure at least one processor to carry out a method according to the first aspect and / or the second aspect and / or said at least one processor to carry out a method for exchanging data according to any one of the embodiments of the present disclosure. The computer program product can use any programming language, and can be in the form of source code, object code, or in any intermediate form between source code and object code, such as in a partially compiled form, or in any other desirable form.

[0079] According to an eighth aspect of the present disclosure, a computer-readable storage medium is provided. The computer-readable storage medium comprises instructions which, when executed by at least one processor, configure said at least one processor to carry out a method according to any one of the embodiments of the presentInternal20250264414disclosure.

[0080] Embodiments of the present disclosure concerns: LCM, model applicability and applicability reporting. The benefits of this approach are a reduced AI / ML performance degradation, and that the network is aware of model inapplicability and configures model parameters more efficiently.

[0081] These and other features, aspects, and advantages will be apparent from a reading of the following detailed description and a review of the associated drawings. It is to be understood that both the foregoing general description and the following detailed description are explanatory and do not restrict aspects as claimed.BRIEF DESCRIPTION OF THE DRAWINGS

[0082] The present disclosure is further described in terms of exemplary embodiments. These exemplary embodiments are described in detail with reference to the drawings. The drawings are not drawn to scale. These embodiments are non-limiting exemplary embodiments, in which like reference numerals represent similar structures throughout the several views of the drawings, and wherein:

[0083] FIG. 1 is a flowchart illustrating an exemplary process performed by the wireless device (UE) side;

[0084] FIG. 2 is a flowchart illustrating an exemplary process performed by the network side (gNB or base station).DETAILED DESCRIPTION

[0085] The present disclosure generally relates to one or both of a system and an apparatus for network connectivity in association with, for example, a wireless device (or User Equipment (UE)) usable for communication. The present disclosure further relates to a method which can be associated with the system and / or the apparatus.

[0086] Generally, energy efficiency and energy savings would be helpful or desired in communication networks. Wireless communication systems are widely deployed to provide various telecommunication services such as telephony, video, data, messaging, broadcasts, or other similar types of services. These wireless communication systems may employ multiple-access technologies capable of supporting communication with multiple users by sharing available system resources with those users (e.g., bandwidth, transmit power, or other resources). Multiple-access technologies can rely on any of code division, time division, frequency division, orthogonal frequency division, single-carrier frequency division, or time division synchronous code division, to name a few. These and otherInternal20250264415multiple access technologies have been adopted in various telecommunication standards to provide a common protocol that enables different wireless devices to communicate on a municipal, national, regional, and even global level.

[0087] Although wireless communication systems have made great technological advancements over many years, challenges still exist. For example, complex and dynamic environments can still attenuate or block signals between wireless transmitters and wireless receivers, undermining various established wireless channel measuring and reporting mechanisms, which are used to manage and optimize the use of finite wireless channel resources. Consequently, there exists a need for further improvements in wireless communications systems to overcome various challenges.

[0088] Together with inapplicability reporting, UE further indicates a simple cause value of inapplicability FFS how to define this simple cause related to model availability and how to capture it in the spec. Upon receiving one or more full inference configuration(s) via RRCReconfiguration message, UE shall maintain all the full inference configuration(s) no matter the full inference configuration is applicable or inapplicable until the network releases it explicitly.

[0089] Due to changes in UE context (e.g., compute resources, memory, battery power) the AI / ML model would not be applicable. The solved by embodiments of the present disclosure is how to avoid the loss of AI / ML-enabled feature or functionality.

[0090] By embodiments of the present disclosure, a predictive module on the UE is defined to pre-evaluate model applicability based on device and network context.

[0091] The detailed description set forth below, with reference to annexed drawings, is intended as a description of various configurations and is not intended to represent the only configurations in which the concepts described herein may be practiced. The detailed description includes specific details for the purpose of providing a thorough understanding of the various concepts. However, it will be apparent to those skilled in the art that these concepts may be practiced without these specific details. In particular, although terminology from 3GPP 5G NR may be used in this disclosure to exemplify embodiments herein, this should not be seen as limiting the scope of the invention.

[0092] Some of the embodiments contemplated herein will now be described more fully with reference to the accompanying drawings. Other embodiments, however, are contained within the scope of the subject matter disclosed herein, the disclosed subject matter should not be construed as limited to only the embodiments set forth herein; rather, these embodiments are provided by way of example to convey the scope of the subject matter to those skilled in the art.Internal20250264416

[0093] Generally, all terms used herein are to be interpreted according to their ordinary meaning in the relevant technical field, unless a different meaning is clearly given and / or is implied from the context in which it is used. All references to a / an / the element, apparatus, component, means, step, etc. are to be interpreted openly as referring to at least one instance of the element, apparatus, component, means, step, etc., unless explicitly stated otherwise. The steps of any methods disclosed herein do not have to be performed in the exact order disclosed, unless a step is explicitly described as following or preceding another step and / or where it is implicit that a step must follow or precede another step. Any feature of any of the embodiments disclosed herein may be applied to any other embodiment, wherever appropriate. Likewise, any advantage of any of the embodiments may apply to any other embodiments, and vice versa. Other objectives, features and advantages of the enclosed embodiments will be apparent from the following description.

[0094] In some embodiments, a more general term “network node” may be used and may correspond to any type of radio network node or any network node, which communicates with a UE (directly or via another node) and / or with another network node. Examples of network nodes are NodeB, MeNB, ENB, a network node belonging to MCG or SCG, base station (BS), multi-standard radio (MSR) radio node such as MSR BS, eNodeB, gNodeB, network controller, radio network controller (RNC), base station controller (BSC), relay, donor node controlling relay, base transceiver station (BTS), access point (AP), transmission points, transmission nodes, RRU, RRH, nodes in distributed antenna system (DAS), core network node (e.g. Mobile Switching Center (MSC), Mobility Management Entity (MME), etc.), Operations & Maintenance (O&M), Operations Support System (OSS), Self Optimized Network (SON), positioning node (e.g. Evolved- Serving Mobile Location Centre (E-SMLC)), Minimization of Drive Tests (MDT), test equipment (physical node or software), etc.

[0095] In some embodiments, the non-limiting term user equipment (UE) or wireless device may be used and may refer to any type of wireless device communicating with a network node and / or with another UE in a cellular or mobile communication system.Examples of UE are target device, device to device (D2D) UE, machine type UE or UE capable of machine to machine (M2M) communication, PDA, PAD, Tablet, mobile terminals, smart phone, laptop embedded equipped (LEE), laptop mounted equipment (LME), USB dongles, UE category M1, UE category M2, ProSe UE, V2V UE, V2X UE, etc.

[0096] Additionally, terminologies such as base station / gNodeB and UE should be considered non-limiting and do in particular not imply a certain hierarchical relation between the two; in general, “gNodeB” could be considered as device 1 and “UE” could beInternal20250264417considered as device 2 and these two devices communicate with each other over some radio channel. And in the following the transmitter or receiver could be either gNodeB (gNB), or UE.

[0097] As will be appreciated by one skilled in the art, aspects of the embodiments may be embodied as a system, apparatus, method, or program product. Accordingly, embodiments may take the form of an entirely hardware embodiment, an entirely software embodiment (including firmware, resident software, micro-code, etc.) or an embodiment combining software and hardware aspects.

[0098] For example, the disclosed embodiments may be implemented as a hardware circuit comprising custom very-large-scale integration (“VLSI”) circuits or gate arrays, off-the-shelf semiconductors such as logic chips, transistors, or other discrete components. The disclosed embodiments may also be implemented in programmable hardware devices such as field programmable gate arrays, programmable array logic, programmable logic devices, or the like. As another example, the disclosed embodiments may include one or more physical or logical blocks of executable code which may, for instance, be organized as an object, procedure, or function.

[0099] Furthermore, embodiments may take the form of a program product embodied in one or more computer readable storage devices storing machine readable code, computer readable code, and / or program code, referred hereafter as code. The storage devices may be tangible, non- transitory, and / or non-transmission. The storage devices may not embody signals. In a certain embodiment, the storage devices only employ signals for accessing code.

[0100] Any combination of one or more computer readable medium may be utilized. The computer readable medium may be a computer readable storage medium. The computer readable storage medium may be a storage device storing the code. The storage device may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, holographic, micromechanical, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing.

[0101] More specific examples (a non-exhaustive list) of the storage device would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random-access memory (“RAM”), a read-only memory (“ROM”), an erasable programmable read-only memory (“EPROM” or Flash memory), a portable compact disc read-only memory (“CD-ROM”), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium thatInternal20250264418can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device.

[0102] Code for carrying out operations for embodiments may be any number of lines and may be written in any combination of one or more programming languages including an object- oriented programming language such as Python, Ruby, Java, Smalltalk, C++, or the like, and conventional procedural programming languages, such as the “C” programming language, or the like, and / or machine languages such as assembly languages. The code may execute entirely on the user’s computer, partly on the user’s computer, as a stand-alone software package, partly on the user’s computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user’s computer through any type of network, including a local area network (“LAN”), wireless LAN (“WLAN”), or a wide area network (“WAN”), or the connection may be made to an external computer (for example, through the Internet using an Internet Service Provider (“ISP”)).

[0103] Furthermore, the described features, structures, or characteristics of the embodiments may be combined in any suitable manner. In the following description, numerous specific details are provided, such as examples of programming, software modules, user selections, network transactions, database queries, database structures, hardware modules, hardware circuits, hardware chips, etc., to provide a thorough understanding of embodiments. One skilled in the relevant art will recognize, however, that embodiments may be practiced without one or more of the specific details, or with other methods, components, materials, and so forth. In other instances, well-known structures, materials, or operations are not shown or described in detail to avoid obscuring aspects of an embodiment. Reference throughout this specification to “one embodiment,” “an embodiment,” or similar language means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment. Thus, appearances of the phrases “in one embodiment,” “in an embodiment,” and similar language throughout this specification may, but do not necessarily, all refer to the same embodiment, but mean “one or more but not all embodiments” unless expressly specified otherwise. The terms “including,” “comprising,” “having,” and variations thereof mean “including but not limited to,” unless expressly specified otherwise. An enumerated listing of items does not imply that any or all of the items are mutually exclusive, unless expressly specified otherwise. The terms “a,” “an,” and “the” also refer to “one or more” unless expressly specified otherwise.

[0104] Aspects of the embodiments are described below with reference to schematicInternal20250264419flowchart diagrams and / or schematic block diagrams of methods, apparatuses, systems, and program products according to embodiments. It will be understood that each block of the schematic flowchart diagrams and / or schematic block diagrams, and combinations of blocks in the schematic flowchart diagrams and / or schematic block diagrams, can be implemented by code. This code may be provided to a processor of a general-purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions / acts specified in the flowchart diagrams and / or block diagrams.

[0105] The code may also be stored in a storage device that can direct a computer, other programmable data processing apparatus, or other devices to function in a particular manner, such that the instructions stored in the storage device produce an article of manufacture including instructions which implement the function / act specified in the flowchart diagrams and / or block diagrams.

[0106] The code may also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus, or other devices to produce a computer implemented process such that the code which execute on the computer or other programmable apparatus provide processes for implementing the functions / acts specified in the flowchart diagrams and / or block diagrams.

[0107] The flowchart diagrams and / or block diagrams in the Figures illustrate the architecture, functionality, and operation of possible implementations of apparatuses, systems, methods, and program products according to various embodiments. In this regard, each block in the flowchart diagrams and / or block diagrams may represent a module, segment, or portion of code, which includes one or more executable instructions of the code for implementing the specified logical function(s).

[0108] It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the Figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. Other steps and methods may be conceived that are equivalent in function, logic, or effect to one or more blocks, or portions thereof, of the illustrated Figures.

[0109] Although various arrow types and line types may be employed in the flowchart and / or block diagrams, they are understood not to limit the scope of the corresponding embodiments. Indeed, some arrows or other connectors may be used to indicate only theInternal20250264420logical flow of the depicted embodiment. For instance, an arrow may indicate a waiting or monitoring period of unspecified duration between enumerated steps of the depicted embodiment. It will also be noted that each block of the block diagrams and / or flowchart diagrams, and combinations of blocks in the block diagrams and / or flowchart diagrams, can be implemented by special purpose hardware-based systems that perform the specified functions or acts, or combinations of special purpose hardware and code.

[0110] The description of elements in each figure may refer to elements of proceeding figures. Like numbers refer to like elements in all figures, including alternate embodiments of like elements.

[0111] The detailed description set forth below, with reference to the figures, is intended as a description of various configurations and is not intended to represent the only configurations in which the concepts described herein may be practiced. The detailed description includes specific details for the purpose of providing a thorough understanding of the various concepts. However, it will be apparent to those skilled in the art that these concepts may be practiced without these specific details. For instance, although 3GPP terminology, from e.g., 5G NR, may be used in this disclosure to exemplify embodiments herein, this should not be seen as limiting the scope of the present disclosure.

[0112] The disclosure is related to wireless communication system, which may be for example a 5G NR wireless communication system. More specifically, it represents a RAN of the wireless communication system, which is used exchange data with UEs via radio signals. For example, the RAN may send data to the UEs (downlink, DL), for instance data received from a core network (CN). The RAN may also receive data from the UEs (uplink, UL), which data may be forwarded to the CN.

[0113] In the examples illustrated, the RAN comprises one base station, BS. Of course, the RAN may comprise more than one BS to increase the coverage of the wireless communication system. Each of these BSs may be referred to as NB, eNodeB (or eNB), gNodeB (or gNB, in the case of a 5G NR wireless communication system), an access point or the like, depending on the wireless communication standard(s) implemented.

[0114] The UEs are located in a coverage of the BS. The coverage of the BS corresponds for example to the area in which UEs can decode a PDCCH transmitted by the BS.

[0115] An example of a wireless device suitable for implementing any method, discussed in the present disclosure, performed at a UE corresponds to an apparatus that provides wireless connectivity with the RAN of the wireless communication system, and that can be used to exchange data with said RAN. Such a wireless device may beInternal20250264421included in a UE. The UE may for instance be a cellular phone, a wireless modem, a wireless communication device, a handheld device, a laptop computer, or the like. The UE may also be an Internet of Things (loT) equipment, like a wireless camera, a smart sensor, a smart meter, smart glasses, a vehicle (manned or unmanned), a global positioning system device, etc., or any other equipment that may run applications that need to exchange data with remote recipients, via the wireless device.

[0116] An example of a wireless device suitable for implementing any method, discussed in the present disclosure, performed at a UE corresponds to an apparatus that provides wireless connectivity with the RAN of the wireless communication system, and that can be used to exchange data with said RAN. Such a wireless device may be included in a UE. The UE may for instance be a cellular phone, a wireless modem, a wireless communication device, a handheld device, a laptop computer, or the like. The UE may also be an Internet of Things (loT) equipment, like a wireless camera, a smart sensor, a smart meter, smart glasses, a vehicle (manned or unmanned), a global positioning system device, etc., or any other equipment that may run applications that need to exchange data with remote recipients, via the wireless device.

[0117] The wireless device comprises one or more processors and one or more memories. The one or more processors may include for instance a central processing unit (CPU), a digital signal processor (DSP), a field-programmable gate array (FPGA), an application specific integrated circuit (ASIC), etc. The one or more memories may include any type of computer readable volatile and non-volatile memories (magnetic hard disk, solid-state disk, optical disk, electronic memory, etc.). The one or more memories may store a computer program product, in the form of a set of program-code instructions to be executed by the one or more processors to implement all or part of the steps of a method for exchanging data, performed at a UE’s side, according to any one of the embodiments disclosed herein.

[0118] The wireless device can comprise also a main radio, MR, unit. The MR unit corresponds to a main wireless communication unit of the wireless device, used for exchanging data with BSs of the RAN using radio signals. The MR unit may implement one or more wireless communication protocols, and may for instance be a 3G, 4G, 5G, NR, Wi-Fi, WiMax, etc. transceiver or the like. In preferred embodiments, the MR unit corresponds to a 5G NR wireless communication unit.

[0119] Embodiments of the present disclosure are relevant to 3GPP Systems (beyond 5G and / or 6G) or Continental / Automotive: AI / ML is an important technology being actively discussed in 3GPP. AI / ML is one of the key topics for Rel-19 in RAN2 AI / ML for airInternal20250264422interface. AI / ML is the next big thing in network optimization, energy efficient networks hence it is very critical to keep up with the topic. Embodiments of the present disclosure are relevant to Rel. 19: AI / ML for air interface (RAN2).

[0120] In 3GPP RAN2 #129 meeting, the following agreement:Agreements on applicability reporting1. Together with inapplicability reporting, UE further indicates a simple cause value of inapplicability FFS how to define this simple cause related to model availability and how we capture it in the spec.2. Upon receiving one or more full inference configuration(s) via RRCReconfiguration message, UE shall maintain all the full inference configuration(s) no matter the full inference configuration is applicable or inapplicable until the network releases it explicitly.3. No prohibit timer is introduced.

[0121] Due to changes in UE context (e.g., compute resources, memory, battery power) the AI / ML model would not be applicable. The problem is how to avoid the loss of AI / ML-enabled feature or functionality.

[0122] FIG. 1 is a flowchart illustrating an exemplary process performed by the wireless device UE side (here: UE side). The method of FIG. 1 is performed by the wireless device for enhancing model applicability in a wireless communication system.

[0123] The user equipment receives configuration information for both a light-weight model and an applicable model from a base station. The configuration information comprises parameters, metadata, or operational details necessary for the execution of at least one artificial intelligence or machine learning model. The user equipment is adapted to determine whether the currently applicable model is inapplicable, for example, due to device context such as compute resources, memory availability, or battery level, or due to network context such as Al model policy, gNB configuration, or slice profile, or due to model metadata such as model size, execution time estimate, or hardware dependency.

[0124] In some embodiments, the applicable model (also referring to primary model) is a designated model which is an AI / ML model designated as primary by the network node, or a full model which is an AI / ML model intended for use under normal device conditions and providing a full feature and / or functionality set. The applicable model may provide a feature and / or functionality set for the AI / ML use case negotiated between the network and the UE. In one case, the network node provides configuration information of Model X to the UE, explicitly designating the Model X as the primary model based on configuration or network policy or a particular use case. In another case, the network provides configuration information of the full model to the UE. The UE utilizes the primary modelInternal20250264423(the full model or the Model X) to perform the AI / ML use cases as long as resources permit. The light-weight model is an alternative AI / ML model intended for use when the applicable model is inapplicable due to device resource constraints, and providing at least part of the feature and / or functionality set of the applicable model.

[0125] For example, the applicable model may be a comprehensive neural network model for channel state prediction. This model utilizes a large number of input features (such as historical channel measurements, mobility patterns, and environmental data) and has a deep architecture, enabling highly accurate and robust predictions. The applicable model is used when the UE has sufficient compute resources, memory, and battery power. The light-weight model may use a reduced set of input features and a shallower network architecture, significantly lowering its computational and memory requirements. Although the light-weight model may provide less accurate predictions, it ensures that the UE can continue to support uplink optimization even under constrained resource conditions.

[0126] If the user equipment determines that the model is inapplicable, it transmits an inapplicability indication to the network, for example, to the base station / network node. The indication can be transmitted via uplink control information, UCI, or uplink MAC control element, UL MAC CE. The determination of model applicability is based on any of UE context, network context, or model metadata. The UE context may comprise compute resources, memory availability, battery level, or other operational parameters. The network context may comprise any of Al model policy, gNB configuration, or slice profile. The model metadata may comprise any of model size, execution time estimate, or hardware dependency. The inapplicability indication comprises a single-bit value, where a value of “1” indicates inapplicability and a value of “0” indicates applicability. The indication is transmitted via uplink control information or uplink MAC control element. Upon reporting the inapplicability indication, the process ends for this flow and waits for response from the base station.

[0127] After reporting, the UE may receive the control information from the network, for example, from the base station / network node. When the UE has received configuration information of both primary and light-weight models, the control information may include switching information for the light-weight model. The control information may include an activation / switching command for the second model. The UE could switch to the lightweight model based on the control information.

[0128] In some embodiments, the UE also takes the UE context into consideration before performing the switch. For example, the UE context changes so that the primary model is inapplicable. The UE sends the inapplicability indication to the base station. AfterInternal20250264424receiving the control information of switching to the light-weight model, the UE may assess the UE context again. If the primary model is still inapplicable with current UE context, the UE could perform the switch. If the UE context improves and the primary model is applicable again, the UE may not perform the switch. The network context and / or the model metadata may be considered in a similar way.

[0129] If the model is determined to be applicable, the user equipment continues operation without reporting inapplicability. In some embodiments, the UE maintains received configuration information, until the network explicitly releases them.

[0130] It should be noted that the figure provides a schematic workflow of the UE, which does not intend to limit the scope of the present disclosure. For example, the UE may first receive configuration information of the primary model. After determining that the primary model is inapplicable in the UE, the UE receives control information from the base station. The control information may include the configuration information of the lightweight model. Alternatively, the control information may include both the configuration information of the light-weight model and a switch command of the light-weight model.

[0131] In some embodiments, before receiving configuration information from the base station, the UE may send a request to the base station to obtain configuration information for one or more models.

[0132] In some embodiments, the UE receives configuration information for a plurality of models to a user equipment (UE). The plurality of models may be designed to address different operating scenarios, such as conserving battery power, reducing memory usage, or optimizing computational efficiency. If the UE determines that the primary model is inapplicable under current conditions, it may transmit an indication to the base station, including a reason for the inapplicability (for example, low battery or insufficient memory). The UE receives control information from the base station indicating selection of a specific model. The UE may maintain all received configurations before receiving control information from the base station.

[0133] In some embodiments, the process would continue when the UE works after changing to the light-weight model. Once the primary model is applicable again in the UE based on the UE context, the UE could switch back from the light-weight model to the primary model. In an example, the UE may perform the switch locally and optionally send a report to the base station. In another example, the UE may send an indication of the primary model applicability status to the base station. The base station could transmit the control information such as a command to instruct the UE to perform the switch. After switching, the UE could report that to the network, for example, to the base station / networkInternal20250264425node.

[0134] The flowchart in FIG. 1 illustrates that the user equipment is adapted to dynamically assess model applicability based on current device context and to efficiently communicate the applicability status to the network, for example, to the base station / network node. This enables the network to respond with appropriate control information or alternative model configurations, thereby reducing performance degradation and improving the flexibility of artificial intelligence and machine learning model management in the wireless communication system.

[0135] FIG. 2 is a flowchart illustrating an exemplary process performed by the network side (gNB or base station). The base station is adapted to provide configuration information for both a light-weight model and an applicable model to the user equipment. The configuration information comprises parameters, operational details, or selection criteria for each model. The light-weight model is adapted for use when the applicable model is determined to be inapplicable due to user equipment context, such as compute resources, memory, or battery level, or due to network context, or due to model metadata.

[0136] The base station determines whether a model inapplicability indication has been received from the user equipment. The model inapplicability indication is a single-bit value, where “0” indicates applicability and “1” indicates inapplicability. The indication is transmitted from the user equipment to the base station via uplink control information or uplink MAC control element.

[0137] If the model inapplicability indication is not received, the base station continues operation without further action. If the model inapplicability indication is received, e.g., the first model applicability status is “inapplicable”, the base station provides an activation or switching indication to the light-weight model and / or provides the light-weight model configuration to the user equipment. The activation or switching indication, or the lightweight model configuration, is transmitted via downlink control information or downlink MAC control element.

[0138] The process ends after the base station provides the activation or switching indication and / or the light-weight model configuration. The base station is further adapted to trigger switching back to the applicable model based on a subsequent indication from the user equipment, indicating that the applicable model is again suitable for operation.

[0139] It should be noted that the figure provides a schematic workflow of the base station, which does not intend to limit the scope of the present disclosure.

[0140] In an example, the base station provides configuration for the primary (applicable) model to the UE firstly. After receiving the indication that the primary model isInternal20250264426inapplicable in the UE, the UE provides control information to the UE. The control information includes configuration of the light-weight model. In addition, the control information may also include switch command for the light-weight model. This avoids consuming UE storage resources.

[0141] In another example, the base station provides configurations for the primary and light-weight models to the UE. After receiving the indication that the primary model is inapplicable in the UE, the UE provides control information to the UE. The control information includes switch command for the light-weight model. This ensures a seamless switch between models.

[0142] In a further example, the base station provides configurations for the primary and light-weight models to the UE. After receiving the indication that the primary model is inapplicable in the UE, the UE provides control information to the UE. The control information includes both the configuration of the light-weight model and the switch command for the light-weight model. This further ensures the availability of the light-weight model.

[0143] In a further example, a base station provides configuration information for a plurality of models to a user equipment (UE). The plurality of models may be designed to address different operating scenarios, such as conserving battery power, reducing memory usage, or optimizing computational efficiency. If the UE determines that the primary model is inapplicable under current conditions, it may transmit an indication to the base station, including a reason for the inapplicability (for example, low battery or insufficient memory). In response, the base station may select, from among the plurality of models, a model that is most suitable for the UE’s current context and provide an activation or switching command to the UE. This enables the UE to dynamically adapt its model usage to changing device conditions, thereby maintaining essential functionality while optimizing resource consumption. Alternatively, the base station may first provide configuration of the primary model to the UE. After receiving the indication, the base station sends the control information including configuration of the suitable model among the plurality models. The embodiments previously described using two models are not limited to two models and are also applicable to multiple models.

[0144] In some embodiments, the UE may send an indication to the base station once the applicability of the primary model changes, for example, from applicable to inapplicable, and vice versa. The base station would provide control information to the UE once receiving an indication. The base station may determine the content of the control information. The content is discussed in the content. For example, the base station hasInternal20250264427sent configuration information of a plurality of models, like both primary and light-weight models. The content of the control information may be a switch command to a specific model

[0145] In an example, the indication sent by the UE may include any of request for specific information, reason of inapplicability. The indication transmitted by the user equipment may include a field or value that explicitly identifies the reason for the inapplicability of the primary model, such as a predefined cause code representing conditions like low battery, insufficient memory, or limited computational resources.

[0146] In some embodiments, after instructing the UE to switch to the light-weight model, the base station may receive a subsequent indication from the UE. The subsequent indication indicates that the first model applicability status is “applicable” in the UE. The base station could transmit further control information to the UE. The further control information may include a switch back command which triggers the UE to switch back from the second model to the first model. In addition, the further control information may include configuration of the primary model to ensure the availability of the primary model.

[0147] Figure 1 shows the UE side flow and in principle the problem is solved by, that the UE and Figure 2 show the network side (gNB) flow.

[0148] The UE receives configurations for both light-weight and applicable, models. Solutions to receive light-weight model configurations:Option #1: light-weight model configuration is received together with applicable model (e.g., via RRCconfiguration).Option #2: whenever UE model is inapplicable, it reports model inapplicability to the network, and network provides light-weight model configuration in response.1 bit is used to indicate model inapplicability to the network ((^applicability,1 inapplicability).Model inapplicability indication is sent through UCI or UL MAC CE.

[0149] Upon receiving model inapplicability indication:Option #1 : Network provides configuration for light-weight model to UE (corresponding to Option #2 of the UE).Option #2: Network activates the light-weight model (which was provided before the activation, for example, the light-weight model was provided initially together with applicable model) through DCI or DL MAC CE (corresponding to Option #1 of the UE).Upon receiving the activation / switching indication from the network, the UE switches to light-weight model.

[0150] Based on UE implementation (e.g., UE context improved, e.g., computeInternal20250264428resources, memory, battery power), the UE provides indication to network to switch from light-weight model to applicable model. The UE provides such indication through UCI or UL MAC CE by indicating “0” (i.e., applicability).

[0151] In this solution, the UE receives configurations for both, light-weight and applicable, models (either light-weight model configuration is received together with applicable model or, whenever UE model is inapplicable, the network provides light-weight model configuration in response to inapplicability indication).

[0152] Aspects of the present disclosure relate to wireless communications, and more particularly, to techniques for configuring the use of a machine learning model in a wireless communication network. Due to changes in UE context (e.g., compute resources, memory, battery power) the AI / ML model would not be applicable. To avoid the loss of AI / ML-enabled feature or functionality, the UE receives configurations for both, light-weight and applicable, models (either light-weight model configuration is received together with applicable model or, whenever UE model is inapplicable, the network provides light-weight model configuration in response to inapplicability indication. The UE performs model switch based on the communication with the base station and its context.

[0153] This solution for pre-empting AI / ML model applicability in a wireless communication network includes standardizing a predictive module on a User Equipment (UE) to pre-evaluate model applicability based on device and network context; the predictive module considering UE context such as CPU load, memory usage, battery level; network state such as Al model policy, gNB configuration, slice profile; model metadata such as size, execution time estimate, hardware dependency; reporting the inference to the network for configuring the UE-side model parameters.

[0154] The Benefits are, that a reduced AI / ML performance degradation is obtained, network is aware of model inapplicability and configures model parameters more efficiently.

[0155] Furthermore, embodiments of the present disclosure relate to 3GPP AI / ML Rel-19 Wl for Air Interface, which could provide mechanisms for improving performance AI / ML, feature / functionality as well as for reducing signaling between radio access network and user devices. Based on the proposed method, AI / ML performance for wireless communications can be enhanced.

[0156] The embodiments below are illustrative and are not intended to limit the scope of the present disclosure.

[0157] The present disclosure solves the cited problem by the proposed embodiments and describes a method for Enhanced Model Applicability in a wireless communicationInternal20250264429system comprising the steps, that a wireless communication unit configured to receive configuration information for a plurality of models; a processor configured to determine applicability of a first model; transmit an indication of model applicability to a network; receive, in response to the activation indication, control information related to a second model; and switch between the first model and the second model based on the control information and UE context.

[0158] In some embodiments of the method according to the first aspect, the method is characterized by, that the plurality of models comprises an applicable model and a lightweight model.

[0159] In some embodiments of the method according to the first aspect, the method is characterized by, that the indication of model applicability comprises a single-bit value indicating inapplicability.

[0160] In some embodiments of the method according to the first aspect, the method is characterized by, that the indication is transmitted via uplink control information (UCI) or uplink MAC control element (UL MAC CE).

[0161] In some embodiments of the method according to the first aspect, the method is characterized by, that the control information comprises switching information for the second model.

[0162] In some embodiments of the method according to the first aspect, the method is characterized by, that the switch between models is based on UE context parameters including at least one of compute resources, memory availability, or battery level.

[0163] According to a second aspect, the present disclosure relates to a method for Enhanced Model Applicability in a wireless communication system comprising the steps, that a wireless communication unit in the wireless communication system configured to transmit configuration information for a plurality of models to a user equipment (UE); a receiver configured to receive from the UE an indication of model applicability; a processor configured to determine, based on the indication, whether to provide control information related to a second model; and transmit the control information to the UE to trigger model switching.

[0164] In some embodiments of the system according to the second aspect, is characterized by, that the control information comprises either a configuration for the second model or an activation command for the second model.

[0165] In some embodiments of the system according to the second aspect, is characterized by, that the control information is transmitted via downlink control information (DCI) or downlink MAC control element (DL MAC CE).Internal20250264430

[0166] In some embodiments of the system according to the second aspect, is characterized by, that the processor is further configured to trigger switching back to the first model based on a subsequent indication from the UE.

[0167] According to a third aspect, the present disclosure relates to an apparatus for network connectivity comprising: a first module configured to receive configuration information for a plurality of models, to transmit an indication of model applicability to a network; and to receive, in response to the activation indication, control information related to a second model; a second module configured to determine applicability of a first model; and to switch between the first model and the second model based on the control information and UE context.

[0168] According to a fourth aspect, the present disclosure relates to an apparatus for network connectivity and is characterized by, that the apparatus corresponds to a User Equipment (UE) communicable with a device corresponding to a base station, and wherein the base station corresponds to a Next generation Node B (gNB) configured to communicate the at least one input signal to the UE.

[0169] According to a fifth aspect, the present disclosure relates to a wireless communication system comprising: at least one apparatus according to the third aspect and at least one device according to the fourth aspect, wherein the apparatus and the device are capable of being coupled via at least one of wired coupling and wireless coupling.

[0170] According to a further aspect, the present disclosure relates to a computer program product comprising instructions which, when executed by at least one processor, configure said at least one processor to carry out a method according to the first aspect and / or said at least one processor to carry out a method for exchanging data according to any one of the embodiments of the present disclosure. The computer program product can use any programming language, and can be in the form of source code, object code, or in any intermediate form between source code and object code, such as in a partially compiled form, or in any other desirable form.

[0171] According to a further aspect, the present disclosure relates to a computer-readable storage medium comprising instructions which, when executed by at least one processor, configure said at least one processor to carry out a method according to any one of the embodiments of the present disclosure.Internal

Claims

20250264431CLAIMS1. A method, performed by a wireless device, for enhancing model applicability in a wireless communication system, the method comprising:receiving configuration information of one or more models from a network node, wherein the one or more models comprise a first model;determining a first applicability status of the first model;transmitting an indication of the first applicability status to the network node; receiving, from the network node, control information related to a second model; andswitching between the first model and the second model based on the control information.

2. The method according to claim 1 , wherein the one or more models comprise a primary model and a light-weight model; andpreferably wherein the first model is the primary model, and the second model is the light-weight model.

3. The method according to claim 2,wherein the primary model is an AI / ML model designated by the network node, or wherein the primary model is an AI / ML model intended for use under normal device conditions and providing a full feature and / or functionality set; andwherein the light-weight model is an alternative AI / ML model intended for use when the primary model is inapplicable due to device resource constraints and providing at least part of the feature and / or functionality set of the primary model.

4. The method according to any of claims 1 to 3, wherein the indication of the first applicability status comprises a single-bit value indicating inapplicability, and / or wherein the indication of the first applicability status is transmitted via uplink control information, UCI, or uplink MAC control element, UL MAC CE; and / orwherein the control information comprises switching information related to the second model, and / orthe indication of the first applicability status comprises a cause of the first applicability status.Internal202502644325. The method according to any of claims 1 to 4, wherein the control information comprises configuration information of the second model.

6. The method according to any of claims 1 to 4, the receiving configuration information of one or more models from the network node comprising:receiving configuration information of the first and second models.

7. The method according to claim 6, wherein the control information comprises an activation / switching command related to the second model.

8. The method according to any of claims 1 to 7, the transmitting an indication of the first applicability status to the network node comprising:transmitting the indication of the first applicability status to the network node when the first model is inapplicable in the wireless device.

9. The method according to any of claims 1 to 8, the determining a first applicability status of a first model comprising:determining the first applicability status based on any of wireless device context, network context, or model metadata, wherein the wireless device context comprises any of compute resources, memory availability, or battery level, wherein the network context comprises any of Al model policy, gNB configuration, or slice profile, and wherein the model metadata comprises any of model size, execution time estimate, or hardware dependency.

10. A method, performed by a network node, for enhancing model applicability in a wireless communication system, the method comprising:transmitting configuration information for one or more models to a wireless device, wherein the one or more models comprise a first model;receiving, from the wireless device, an indication of a first model applicability status; determining, based on the indication, whether to provide control information related to a second model; andtransmitting the control information to the wireless device to trigger model switching.Internal2025026443311. The method according to claim 10, wherein the control information comprises configuration information of the second model and / or an activation / switching command related to the second model; and / orwherein the control information is transmitted via downlink control information, DCI, or downlink MAC control element, DL MAC CE; and / orthe network node corresponds to a Next generation Node B, gNB, configured to communicate the at least one input signal to the wireless device.

12. The method according to claims 10 or 11 , further comprising:receiving, from the wireless device, a subsequent indication, wherein the subsequent indication indicates that the first model is applicable in the wireless device; transmitting a switch back command to the wireless device, wherein the switch back command triggers the wireless device to switch back from the second model to the first model.

13. A wireless device for enhancing model applicability in a wireless communication system, comprising:a communication module configured to:receive configuration information for one or more models from a network node, wherein the one or more models comprise a first model;transmit an indication of a first model applicability status to a network node; andreceive, from the network node, control information related to a second model;a processing module configured to:determine the first applicability status of the first model; andswitch between the first model and the second model based on the control information.

14. A network node for enhancing model applicability in a wireless communication system, comprising:a communication module configured to:transmit configuration information for one or more models to a wireless device, wherein the one or more models comprise a first model;Internal20250264434receive, from the wireless device, an indication of a first model applicability status; andtransmit control information to the wireless device to trigger model switching; a processing module configured to:determine, based on the indication, whether to provide control information related to a second model.

15. A wireless communication system for enhancing model applicability comprising: at least one wireless device according to any of claims 1 to 9; andat least one network node according to any of claims 10 to 12,wherein the at least one wireless device and the at least one network node are capable of being coupled via at least one of wired coupling and / or wireless coupling.Internal