Ai / ML model switching during RRC re-establishment
Patent Information
- Application Number
- PCT/EP2026/058624
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2025-03-28
- Filing Date
- 2026-03-25
- Publication Date
- 2026-10-01
Smart Images
Figure EP2026058624_01102026_PF_FP_ABST
Abstract
Description
[0001] 202501721
[0002] - 1 -
[0003] TITLE
[0004] Method and system for AI / ML model switching during RRC Re-establishment
[0005] TECHNNICAL FIELD
[0006] The present invention disclosure relates to wireless communication networks, specifically to AI / ML model switching mechanisms during RRC (radio resource control) re-establishment in 3GPP (3rd generation partnership project) RAN (radio access network) environments. The invention provides an efficient method to reduce model switching latency, signaling overhead, and inference disruption during AI / ML model-based network operations.
[0007] BACKGROUND
[0008] In 3GPP (Third Generation Partnership Project), one of the selected study items as the approved Release 18 package is AI / ML (artificial intelligence / machine learning) as described in the related document (RP-213599) addressed in 3GPP TSG (Technical Specification Group) RAN (Radio Access Network) meeting #94e. The official title of AI / ML study item is “Study on AI / ML for NR Air Interface”. The goal of this study item is to identify a common AI / ML framework and areas of obtaining gains using AI / ML based techniques with use cases. According to 3GPP, the main objective of this study item is to study AI / ML framework for air-interface with target use cases by considering performance, complexity, and potential specification impact. In particular, AI / ML model, terminology and description to identify common and specific characteristics for framework are included as one of key work scopes.
[0009] Regarding AI / ML framework, various aspects are under consideration for investigation and one of key items is about lifecycle management of AI / ML model where multiple stages are included as mandatory for model training, model deployment, model inference, model monitoring, model updating etc. Also in 3GPP, two-sided (AI / ML) model is defined as a paired AI / ML model(s) over which joint inference is performed, where joint inference comprises AI / ML Inference whose inference is performed jointly across the UE and the network. Also for one-sided202501721
[0010] -2 -
[0011] (AI / ML) model, UE-side (AI / ML) model is defined as an AI / ML model whose inference is performed entirely at the UE and network-side (AI / ML) model is defined as an AI / ML model whose inference is performed entirely at the network. Currently, AI / ML specification work is at the stage of work item discussion for Release 19.
[0012] Earlier, in 3GPP TR 37.817 for Release 17, titled as Study on enhancement for Data Collection for NR and EN-DC, UE (user equipment) mobility was also considered as one of AI / ML use cases and one of scenarios for model training / inference is that both functions are located within RAN node. Followingly, in Release 18 the new work item of “Artificial Intelligence (AI)ZMachine Learning (ML) for NG-RAN” was initiated to specify data collection enhancements and signaling support within existing NG-RAN interfaces and architecture. For the above active standardization works, RAN-based AI / ML model is considered very significant for both network and UE to meet any desired model operations (e.g., model training, inference, selection, switching, update, monitoring, etc.).
[0013] Model information can be signaled to pair both network-side and UE-side models for various lifecycle management (LCM) operations. However, signaling overhead indicating model information can be very high especially when model based LCM is processed between base station (BS / gNB) and multiple UEs. In LCM, model training is one of the most important parts for model deployment and currently there is no specification defined for signaling methods and network-UE behaviors so as to identify the required dataset when model updating / re-training as any activated model can be also impacted due to model / data drift. When ML condition changes, the enabled AI / ML model(s) can be impacted for model performance due to data / model drift. In this case, model re-training / updating can be executed.
[0014] This application proposes a novel unified model identification framework to ensure efficient model deployment, transfer, and management. The existing 3GPP framework does not support a conditional, pre-configured AI / ML model switching mechanism, which can dynamically adapt to radio conditions, computational resources, and inference quality. Traditional AI / ML model deployment often lacks standardized compatibility across RAN, core networks (CN), and inter-operator202501721
[0015] - 3 -
[0016] environments, leading to inefficiencies in model execution, synchronization, and adaptation. There are several functional units defined in RAN such as Rll (radio unit), DU (distributed unit), CU (central unit) as well as TRP (transmission reception point) as a network node. In conventional cellular networks, the association between a user equipment (UE) and a base station (BS) is based on geographic cell boundaries, limiting flexibility.
[0017] In RAN, signaling is crucial for communication between the UE and the network as this signaling occurs across different layers of the protocol stack, primarily L1 (Layer 1), L2 (Layer 2), and RRC (radio resource control). Current approaches lack standardized identifiers and signaling procedures to support AI / ML-driven associations, leading to inefficiencies in model synchronization and adaptation.
[0018] Future Al-native networks will integrate AI / ML models at both the UE level and network level to enable intelligent radio access, autonomous operation, and enhanced user experience. In 3GPP-defined AI / ML model deployment scenarios, UE may utilize AI / ML models for various network optimization tasks such as beam selection, resource allocation, and mobility prediction.
[0019] When switching from an prior AI / ML model to a new one, the UE must typically reinitialize model parameters, leading to increased signaling overhead and delay. Moreover, during RRC re-establishment, existing RRC signaling does not account for AI / ML model switching, which introduces additional delay and network inefficiency when the UE needs to transition to a new model. Existing solutions do not efficiently utilize historical AI / ML model data, leading to suboptimal model transition, increased inference latency, and excessive signaling exchanges between UE and gNB.
[0020] Therefore, an improved method is needed to enable seamless AI / ML model switching with minimal performance degradation and signaling cost. RRC reestablishment is a critical procedure in wireless communication systems, particularly in scenarios involving radio link failures, handovers, or UE capability. With the increasing integration of AI / ML models in 5G and beyond networks, there is a need for efficient methods to switch UE-sided AI / ML models during RRC re-establishment while minimizing associated signaling overhead and switching delays. The present202501721
[0021] -4 -
[0022] invention disclosure introduces a novel model switching / activation to address these issues efficiently.
[0023] US 2023069342 describes how to assist determination of the model update time in consideration of cost for the update of a model.
[0024] US 2023022737 explains supporting generation of machine learning model when a certain machine learning model is changed.
[0025] US 2019012876 provides projections, predictions, and recommendations for computing system.
[0026] US 2019332895 shows that the monitored states are to decide to change a trained ML model as currently used.
[0027] EP 4075348 describes control of machine learning model, which can be based on a federated learning method collectively performed by nodes of a decentralized distributed database.
[0028] US 2021019612 provides the self-healing system that can automatically provide a diagnostic, and it can also automatically provide an action if the performance of the model predictions has changed over time.
[0029] The present disclosure solves the cited problem by the proposed embodiments and describes as first aspect a method for AI / ML model switching integrated with RRC re-establishment in a wireless communication system, comprising detecting, by the UE, an RRC re-establishment trigger due to link failure, handover failure, or UE failure while operating an AI / ML model; transmitting, by the UE, an RRCConnectionReestablishmentRequest message including a ModelContextTransfer IE; receiving, by the network node, the RRCConnectionReestablishmentRequest and preloading the new AI / ML model using prior model insights from the ModelContextTransfer IE; transmitting, by the network node, an RRCConnectionReestablishment message including model activation202501721
[0030] - 5 -
[0031] parameters; and activating, by the UE, the new AI / ML model upon RRC reestablishment.
[0032] In some embodiments of the method according to the first aspect the method is characterized by, that the ModelContextTransfer IE comprises a new AI / ML model identifier; a prior AI / ML model identifier; a compressed model attribute data for optimizing model switching; a priority indicator for model selection; and a switching mode indicator defining transition behavior.
[0033] In some embodiments of the method according to the first aspect the method is characterized by, that the new AI / ML model ID is used to identify the candidate AI / ML models available for activation by including a list of model IDs representing potential replacements for the current model.
[0034] In some embodiments of the method according to the first aspect the method is characterized by, that the prior AI / ML model ID is used to identify the AI / ML model currently in use, which is also due for replacement.
[0035] In some embodiments of the method according to the first aspect the method is characterized by, that the switching mode indicator having three operational modes in the ModelContextTransfer IE comprises prior model continuation mode (e.g., the UE continues to operate the prior model even after RRC re-establishment), hybrid operation mode (e.g., the UE executes a lightweight version of the new model alongside the prior model while transitioning); and immediate switch mode (e.g., the UE activates the new model as soon as RRC re-establishment is completed).
[0036] In some embodiments of the method according to the first aspect the method is characterized by, that the ModelContextTransfer IE delivered via UEAssistancelnformation includes a priority indicator that specifies a ranked list of AI / ML models for switching.
[0037] In some embodiments of the method according to the first aspect the method is characterized by, that the compressed model attribute data used for compact202501721
[0038] -6 -
[0039] representation of critical model characteristics to facilitate efficient switching in the ModelContextTransfer IE comprises: {model type, model size, model version, model parameters}; {model input feature data}; {model lifecycle}; and {model performance metric}.
[0040] In some embodiments of the method according to the first aspect the method is characterized by, that the priority indicator is used to identify the priority level assigned to each AI / ML model to guide selection by ensuring the most suitable model is chosen under varying conditions.
[0041] The present disclosure solves the cited problem by the proposed embodiments and describes as second aspect a method for proactive AI / ML model switching with RRC re-configuration during RRC Connected state in a wireless communication system, comprising .transmitting, by the UE, UEAssistancelnformation including ModelContextTransfer IE to notify the network of the predicted model transition; Receiving, by the network node, the UEAssistancelnformation and preparing AI / ML model switching parameters; transmitting, by the network node, an RRCReconfiguration message containing model activation instructions; and activating, by the UE, the preloaded AI / ML model while maintaining network connection.
[0042] The present disclosure solves the cited problem by the proposed embodiments and describes as third aspect as an apparatus for AI / ML model switching integrated with RRC re-establishment in a wireless communication system the apparatus comprising a wireless transceiver, a processor coupled with a memory in which computer program instructions are stored, said instructions being configured to carry out the steps according to the first and second aspect.
[0043] The present disclosure solves the cited problem by the proposed embodiments and describes as fourth aspect as a user Equipment comprising an apparatus according to third aspect202501721
[0044] -7 -
[0045] The present disclosure solves the cited problem by the proposed embodiments and describes as fifth aspect as a gNB comprising an apparatus according to third aspect.
[0046] The present disclosure solves the cited problem by the proposed embodiments and describes as sixth aspect a wireless communication system for AI / ML model switching integrated with RRC re-establishment, wherein the wireless communication systems comprises at least a user equipment according fourth aspect, at least a gNB according to the fifth aspect, whereby the user Equipment and the gNB each comprises a processor coupled with a memory in which computer program instructions are stored, said instructions being configured to carry out the steps according to the first and second aspect.
[0047] According to a seventh 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 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.
[0048] According to a eighth 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.
[0049] SUMMARY OF THE INVENTION
[0050] The present invention disclosure provides a method and system for integrating AI / ML model switching into the RRC re-establishment process, thereby reducing model switching latency and ensuring uninterrupted AI / ML model operations. By modifying the existing RRCReestablishmentRequest message to include a compact AI / ML202501721
[0051] - 8 -
[0052] model identification information, the invention enables efficient model switching during re-establishment procedures.
[0053] BRIEF DESCRIPTION OF THE DRAWINGS
[0054] Figure 1 is an exemplary signaling flow of AI / ML model switching during RRC Reestablishment.
[0055] Figure 2 is an exemplary flow chart of AI / ML model switching process during RRC Re-establishment.
[0056] Figure 3 is an exemplary signaling flow of AI / ML model switching during RRC Connected state.
[0057] Figure 4 is an exemplary table of field information of ModelContextTransfer IE.
[0058] DETAILED DESCRIPTION
[0059] 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.
[0060] 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.
[0061] 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 given202501721
[0062] - 9 -
[0063] 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.
[0064] 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.
[0065] 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 mounted202501721
[0066] - 10 -
[0067] equipment (LME), USB dongles, UE category Ml, UE category M2, ProSe UE, V2V UE, V2X UE, etc.
[0068] 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 be considered 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.
[0069] 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.
[0070] 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.
[0071] 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 code202501721
[0072] - 11 -
[0073] 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.
[0074] 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 readonly 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 that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device.
[0075] 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”)).
[0076] Furthermore, the described features, structures, or characteristics of the embodiments may be combined in any suitable manner. In the following description,202501721
[0077] - 12 -
[0078] 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.
[0079] Aspects of the embodiments are described below with reference to schematic flowchart 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 diagrams202501721
[0080] - 13 -
[0081] 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.
[0082] 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.
[0083] 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).
[0084] 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.
[0085] 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 the logical flow of the depicted embodiment. For instance, an arrow may indicate a waiting or monitoring period of unspecified duration between enumerated steps of the202501721
[0086] - 14 -
[0087] 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.
[0088] 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.
[0089] 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.
[0090] 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.
[0091] 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)202501721
[0092] - 15 -
[0093] implemented.
[0094] 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.
[0095] 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.
[0096] 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 programcode 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.
[0097] 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, WiFi, WiMax, etc. transceiver or the like. In preferred embodiments,202501721
[0098] - 16 -
[0099] the MR unit corresponds to a 5G NR wireless communication unit.
[0100] AI / ML Model is a data driven algorithm that applies AI / ML techniques to generate set of outputs based on set of inputs.
[0101] AI / ML model delivery is a generic term referring to delivery of an AI / ML model from one entity to another entity in any manner. Note is An entity could mean network node / function (e.g., gNB, LMF, etc.), UE, proprietary server, etc.
[0102] AI / ML model Inference is a process of using trained AI / ML model to produce set of outputs based on set of inputs.
[0103] AI / ML model testing is a subprocess of training, to evaluate the performance of final AI / ML model using dataset different from one used for model training and validation. Differently from AI / ML model validation, testing does not assume subsequent tuning of the model.
[0104] AI / ML model training is a process to train an AI / ML Model [by learning the input / output relationship] in data driven manner and obtain the trained AI / ML Model for inference.
[0105] AI / ML model transfer is a delivery of an AI / ML model over the air interface in manner that is not transparent to 3GPP signaling, either parameters of model structure known at the receiving end or new model with parameters. Delivery may contain full model or partial model.
[0106] AI / ML model validation is a subprocess of training, to evaluate the quality of an AI / ML model using dataset different from one used for model training, that helps selecting model parameters that generalize beyond the dataset used for model training.
[0107] Data collection is a process of collecting data by the network nodes, management entity, or UE for the purpose of AI / ML model training, data analytics and inference.202501721
[0108] - 17 -
[0109] Federated learning I federated training is a machine learning technique that trains an AI / ML model across multiple decentralized edge nodes e.g., UEs, gNBs each performing local model training using local data samples. The technique requires multiple interactions of the model, but no exchange of local data samples.
[0110] Functionality identification is a process / method of identifying an AI / ML functionality for the common understanding between the NW and the UE. Note is Information regarding the AI / ML functionality may be shared during functionality identification. Where AI / ML functionality resides depends on the specific use cases and sub use cases.
[0111] Model activation means enable an AI / ML model for specific AI / ML-enabled feature.
[0112] Model deactivation means disable an AI / ML model for specific AI / ML-enabled feature.
[0113] Model download means Model transfer from the network to UE.
[0114] Model identification is A process / method of identifying an AI / ML model for the common understanding between the NW and the UE. The process / method of model identification may or may not be applicable and regarding the AI / ML model may be shared during model identification.
[0115] Model monitoring is A procedure that monitors the inference performance of the AI / ML model.
[0116] Model parameter update is Process of updating the model parameters of model. Model selection is the process of selecting an AI / ML model for activation among multiple models for the same AI / ML enabled feature. Model selection may or may not be carried out simultaneously with model activation.
[0117] Model switching is deactivating currently active AI / ML model and activating different AI / ML model for specific AI / ML-enabled feature.202501721
[0118] - 18 -
[0119] Model update is Process of updating the model parameters and / or model structure of model.
[0120] Model upload is Model transfer from UE to the network.
[0121] Network-side (AI / ML) model is an AI / ML Model whose inference is performed entirely at the network.
[0122] Offline field data is the data collected from field and used for offline training of the AI / ML model.
[0123] Offline training is an AI / ML training process where the model is trained based on collected dataset, and where the trained model is later used or delivered for inference. Note is This definition only serves as guidance. There may be cases that may not exactly conform to this definition but could still be categorized as offline training by commonly accepted conventions.
[0124] Online field data is the data collected from field and used for online training of the AI / ML model.
[0125] Online training is an AI / ML training process where the model being used for inference) is (typically continuously) trained in (near) real-time with the arrival of new training samples. Note is the notion of (near) real-time vs. non real-time is context-dependent and is relative to the inference time-scale. This definition only serves as guidance.
[0126] There may be cases that may not exactly conform to this definition but could still be categorized as online training by commonly accepted conventions. Note is Fine-tuning / re-training may be done via online or offline training. This note could be removed when we define the term fine-tuning.
[0127] Reinforcement Learning (RL) is a process of training an AI / ML model from input (a.k.a. state) and feedback signal (a.k.a. reward) resulting from the model’s output (a.k.a. action) in an environment the model is interacting with.202501721
[0128] - 19 -
[0129] Semi-supervised learning is a process of training model with mix of labelled data and unlabelled data.
[0130] Supervised learning is a process of training model from input and its corresponding labels.
[0131] Two-sided (AI / ML) model is a paired AI / ML Model(s) over which joint inference is performed, where joint inference comprises AI / ML Inference whose inference is performed jointly across the UE and the network, i.e, the first part of inference is firstly performed by UE and then the remaining part is performed by gNB, or vice versa.
[0132] UE-side (AI / ML) model is an AI / ML Model whose inference is performed entirely at the UE.
[0133] Unsupervised learning is a process of training model without labelled data.
[0134] Proprietary-format models is ML models of vendor-Zdevice-specific proprietary format, from 3GPP perspective. They are not mutually recognizable across vendors and hide model design information from other vendors when shared.
[0135] Open-format models is ML models of specified format that are mutually recognizable across vendors and allow interoperability, from 3GPP perspective. They are mutually recognizable between vendors and do not hide model design information from other vendors when shared.
[0136] The following explanation will provide the detailed description of the mechanism about pre-configuring and signaling the specific information about model online training by configuring a set of UE behaviors. AI / ML based techniques are currently applied to many different applications and 3GPP also started to work on its technical investigation to apply to multiple use cases based on the observed potential gains. AI / ML lifecycle can be split into several stages such as data collection / pre-202501721
[0137] - 20 -
[0138] processing, model training, model testing / validation, model deployment / update, model monitoring etc., where each stage is equally important to achieve target performance with any specific model(s). In applying AI / ML model for any use case or application, one of the challenging issues is to manage the lifecycle of AI / ML model. It is mainly because the data / model drift occurs during model deployment and inference and it results in performance degradation of AI / ML model. Fundamentally, the dataset statistical changes occur after model is deployed and model inference capability is also impacted with unseen data as input. In a similar aspect, the statistical property of dataset and the relationship between input and output for the trained model can be changed with drift occurrence. In this context, model training or re-training is one of key issues for model performance maintenance as model performance such as inferencing and / or training is dependent on different model execution environment with varying configuration parameters.
[0139] To handle this issue, collaboration between UE and gNB is highly important to track model performance and re-configure model corresponding to different environments. AI / ML model needs model monitoring after deployment because model performance cannot be maintained continuously due to drift and update feedback is then provided to re-train / update the model or select alternative model. When AI / ML model enabled wireless communication network is deployed, it is then important to consider how to handle AI / ML model in activation with re-configuration for wireless devices under operations such as model training, inference, updating, etc. For determining information about model identification (e.g., model ID), frequent model ID assignment / re-assignment processes might occur due to model drift related to model performance variation and / or model applicable condition change.
[0140] In addition, if all identified models at UE side are to be assessed and monitored, the associated signaling overhead and computing power demand can increase significantly. And signaling overhead can highly increase for model ID decision and assignment. When there are multiple models for model transfer / delivery, significant increase of signaling overhead and the limited UE capability supporting models to be transferred can be critical for the deployment of target models at UE side. In this method, AI / ML model switching is integrated with RRC Re-establishment procedure202501721
[0141] - 21 -
[0142] in radio access networks. Specifically, UE first detects an RRC re-establishment trigger (link failure, handover failure, or UE failure) and include the new information (e.g., new proposed information element such as “ModelContextTransfer IE”) to an RRCConnectionReestablishmentRequest message such as {new AI / ML model ID, prior model ID, compressed model attribute data, priority indicator, switching mode indicator} along with the existing information delivered in this message (e.g., old C-RNTI and old PCI). The gNB then receives the request and uses ModelContextTransfer IE to preload the new AI / ML model with fast UE context recovery by using insights from the prior model’s operational data. Sequentially, the gNB responds with RRCConnectionReestablishmentResponse, including preloaded model activation parameters so that the UE switches to the new model by minimizing model transition impact. The ModelContextTransfer IE transfers key AI / ML model information to ensure seamless model switching without reloading the entire model and also provides essential prior model inference data for fast new model adaptation.
[0143] If RRC re-establishment is triggered, the gNB already prepares preloaded model parameters in advance, reducing model switching latency. Before initiating the model switch, the UE maintains a lightweight buffer of key operational metrics from the prior model, such as feature embeddings, statistical distributions, and intermediate inference outputs. These metrics are stored temporarily within the UE device memory. Instead of activating the new model only after the prior model fails, a lightweight version of the new model (or its initial weights) is preloaded based on predictive triggers on UE side and the network (gNB) can provide proactive indications about expected model switches due to network conditions via reception of UE assistance information that is configured to deliver ModelContextTransfer IE alternatively, enabling the UE to pre-cache the new model. Otherwise, the UE does not immediately halt operations, instead, it continues using the prior model while asynchronously integrating the new model when a model switch occurs due to link failure or handover failure.
[0144] Regarding ModelContextTransfer IE, the field of switching mode indicator has three modes of prior model continuation, hybrid operation, and immediate switch. Among three modes in the field of switching mode indicator, prior model is maintained to202501721
[0145] - 22 -
[0146] operate without model switching even after RRC re-estabalishment when mode of prior model continuation is set. And if the hybrid operation mode is set, a lightweight version of the new on-device model is used with or without prior model operation while seamlessly transitioning to the new model via reception of RRCConnectionReestablishmentResponse. And if the immediate switch mode is set, the UE activates the new model as soon as the RRC connection is restored while prior model operation continues until model switching. The field of priority indicator represents a list of the prioritized models for model switching, especially when there are multiple target models for switching. The field of compressed model attribute data indicate AI / ML model-specific properties (e.g., condensed feature map extracted from the prior model, allowing the new model to maintain continuity without full retraining, or features most relevant in the prior model) that helps the AI / ML model switching seamlessly during an RRC re-establishment. The specific information about compressed model attribute data depends on deployment scenarios, model applications, and model transitions, etc.
[0147] In summary, there are two proposed ways of delivering ModelContextTransfer IE such as UE assistance information (e.g., sent via UEAssistancelnformation before model switching to proactively notify gNB if necessary) and RRC Re-establishment request (e.g., sent via RRCReestablishmentRequest to transfer necessary prior model insights before RRC re-establishment completes). To support a predictive model pre-selection that operates in conjunction with the RRC Re-establishment procedure, the UE's prediction triggers the model pre-selection process by detecting conditions to enable RRC Re-establishment (e.g., deteriorating radio link quality). The UE then evaluates available on-device models in its registry based on current network conditions and selects the most appropriate one and the chosen model is pre-loaded into a standby cache, ready for rapid activation. When RRC Reestablishment is initiated, the UE includes the pre-selected model's identifier in the enhanced RRCReestablishmentRequest message. If the network accepts the reestablishment, it uses the received model identifier to prepare for the new AI / ML configuration as the network signals the activation of the pre-selected AI / ML model so that the UE activates the pre-loaded model from its cache, minimizing switching delay.202501721
[0148] - 23 -
[0149] Figure 1 shows an exemplary signaling flow of AI / ML model switching during RRC Re-establishment. In this example, UE initiates RRC Re-establishment request as the UE detects an RRC failure (e.g., handover failure, radio link failure). To recover the connection, the UE transmits an RRCConnectionReestablishmentRequest message to the network. This request includes the ModelContextTransfer IE, which contains prior AI / ML model ID (currently in use before failure), new AI / ML model ID (candidate model for activation), compressed model attribute data (essential parameters for efficient model switching). Network determines model switching parameters. Upon receiving the RRCConnectionReestablishmentRequest, the network (gNB) extracts the ModelContextTransfer IE to determine whether a model switch is necessary. The key parameters such as network conditions, model performance metrics, and transition confidence score are evaluated to decide the optimal model switching. After then, network responds with RRC Re-establishment message by transmitting RRCConnectionReestablishment message to the UE, including model switching parameters, specifying model context by allowing the new AI / ML model to be activated. UE then activates a new model and completes RRC recovery, processes the received parameters, and activates the new AI / ML model accordingly. After successfully applying the new model and re-establishing the RRC connection, the UE transmits an RRCConnectionReestablishmentComplete message.
[0150] Figure 2 shows an exemplary flow chart of AI / ML model switching process during RRC Re-establishment. In this example, an RRC re-establishment event occurs due to link failure (e.g., poor signal conditions, interference), handover failure (e.g., unsuccessful cell transition), or UE failure (e.g., internal processing issue affecting connectivity). At this point, the AI / ML model is in operation, and a transition may be required. UE then sends RRCConnectionReestablishmentRequest with ModelContextTransfer IE. UE transmits RRCConnectionReestablishmentRequest message to the gNB and this message includes a ModelContextTransfer IE. gNB processes ModelContextTransfer IE and prepares model switching parameters. Upon receiving the RRCConnectionReestablishmentRequest, the gNB extracts model transition data from ModelContextTransfer IE, determines appropriate model202501721
[0151] -24 -
[0152] switching parameters, and prepares necessary adjustments for AI / ML model activation in coordination with RRC recovery. gNB sends RRCConnectionReestablishment with model switching parameters. The gNB transmits an RRCConnectionReestablishment message to the UE. UE completes RRC Re-establishment and activates new model. To confirm successful completion, the UE sends an RRCConnectionReestablishmentComplete message to the gNB.
[0153] Figure 3 shows an exemplary signaling flow of AI / ML model switching during RRC Connected state. In this example, UE sends assistance information with ModelContextTransfer IE. The UE transmits a UE assistance information message to the network, embedding the ModelContextTransfer IE. This IE contains current AI / ML model details (e.g., model ID, version, and type), recommended new model(s) based on real-time network conditions, compressed model attributes, and indications of network or UE constraints influencing model selection while any combinations (or subset) of field elements are allowed to be configured for ModelContextTransfer IE. The ModelContextTransfer IE delivered via UEAssistancelnformation also includes a priority indicator that specifies a ranked list of AI / ML models for switching. The gNB processes the received ModelContextTransfer IE and extracts UE's reported model transition data, AI / ML model compatibility details, resource requirements for the new model. Network then determines model switching parameters. The gNB transmits an RRCReconfiguration message to the UE, containing updated AI / ML model activation instructions, fine-tuned model parameters optimized for network efficiency, and any required radio resource adjustments to support the model switch. UE applies RRC Reconfiguration and activates new model. The UE processes the received model switching parameters. It applies the RRC reconfiguration settings while activating the new AI / ML model.
[0154] Figure 4 shows an exemplary table of field information of ModelContextTransfer IE. In this example, the fields of a new information element within the ModelContextTransfer IE are defined, which enables seamless AI / ML model switching in network operations. For new AI / ML model ID field, it is used to identify the candidate AI / ML models available for activation by including a list of model IDs representing potential replacements for the current model. For prior AI / ML model ID202501721
[0155] - 25 -
[0156] field, the AI / ML model currently in use is specified, which is also due for replacement. This field helps track the model transition process by referencing the old model ID and enables the network to compare previous and new models for performance optimization. For compressed model attribute data field, it provides a compact representation of critical model characteristics to facilitate efficient switching.
[0157] This field is structured into three subcategories such as {model type, model size, model version, model parameters}, {model input feature data}, {model lifecycle}, and {model performance metric}, where {model type, model size, model version, model parameters} defines the architectural model information, {model input feature data} summarizes the feature set the model relies on for inference or training, {model lifecycle} indicates a specific stage of LCM such as model training or model inferencing, and {model performance metric} includes indicators such as accuracy, latency, and efficiency, helping the optimal model to be selected. For priority indicator field, it is used to identify the priority level assigned to each AI / ML model to guide selection by ensuring the most suitable model is chosen under varying conditions.
[0158] For switching mode indicator field, it is used to define the transition between AI / ML models where there are three switching modes such as prior model continuation (e.g., the UE continues using the existing model if switching is unnecessary), hybrid operation (e.g., both the old and new models run in parallel for a smooth transition), immediate switch (e.g., the UE instantly shifts to the new model, replacing the old one without a transition phase). Any combinations (or subset) of field elements are allowed to be configured for ModelContextTransfer IE. The proposed method above contributes to reduction of AI / ML model switching latency by leveraging the existing RRC signaling along with minimization of model switching performance impact.
Claims
202501721- 26 -CLAIMS1. A method for AI / ML model switching integrated with RRC re-establishment in a wireless communication system, comprising:• Detecting, by the UE, an RRC re-establishment trigger due to link failure, handover failure, or UE failure while operating an AI / ML model;• Transmitting, by the UE, an RRCConnectionReestablishmentRequest message including a ModelContextTransfer IE;• Receiving, by the network node, the RRCConnectionReestablishmentRequest and preloading the new AI / ML model using prior model insights from the ModelContextTransfer IE; • Transmitting, by the network node, an RRCConnectionReestablishment message including model activation parameters; and• Activating, by the UE, the new AI / ML model upon RRC re-establishment.
2. The method according to claim 1 , wherein the ModelContextTransfer IE comprises:• A new AI / ML model identifier;• A prior AI / ML model identifier;• Compressed model attribute data for optimizing model switching;• A priority indicator for model selection; and• A switching mode indicator defining transition behavior.
3. The method according to claim 2, wherein the new AI / ML model ID is used to identify the candidate AI / ML models available for activation by including a list of model IDs representing potential replacements for the current model.
4. The method according to claim 2, wherein the prior AI / ML model ID is used to identify the AI / ML model currently in use, which is also due for replacement.
5. The method according to claim 2, wherein the switching mode indicator having three operational modes in the ModelContextTransfer IE comprises:202501721- 27 -• Prior model continuation mode (e.g., the UE continues to operate the prior model even after RRC re-establishment);• Hybrid operation mode (e.g., the UE executes a lightweight version of the new model alongside the prior model while transitioning); and• Immediate switch mode (e.g., the UE activates the new model as soon as RRC re-establishment is completed).
6. The method according to claim 5, wherein the ModelContextTransfer IE delivered via UEAssistancelnformation includes a priority indicator that specifies a ranked list of AI / ML models for switching..
7. The method according to claim 2, wherein the compressed model attribute data used for compact representation of critical model characteristics to facilitate efficient switching in the ModelContextTransfer IE comprises:• {model type, model size, model version, model parameters};• {model input feature data};• {model lifecycle}; and• {model performance metric}.
8. The method according to claim 2, wherein the priority indicator is used to identify the priority level assigned to each AI / ML model to guide selection by ensuring the most suitable model is chosen under varying conditions.
9. A method for proactive AI / ML model switching with RRC re-configuration during RRC Connected state in a wireless communication system, comprising:• Transmitting, by the UE, UEAssistancelnformation including ModelContextTransfer IE to notify the network of the predicted model transition;• Receiving, by the network node, the UEAssistancelnformation and preparing AI / ML model switching parameters;• Transmitting, by the network node, an RRCReconfiguration message containing model activation instructions; and202501721- 28 -Activating, by the UE, the preloaded AI / ML model while maintaining network connection.
10. Apparatus for AI / ML model switching integrated with RRC re-establishment in a wireless communication system the apparatus comprising a wireless transceiver, a processor coupled with a memory in which computer program instructions are stored, said instructions being configured to implement steps of the claims 1 to 9.
11. User Equipment comprising an apparatus according to claim 10.12.gNB comprising an apparatus according to claim 10.
13. Wireless communication system for AI / ML model switching integrated with RRC re-establishment, wherein the wireless communication systems comprises at least a user equipment according to claim 11 , at least a gNB according to claim 12, whereby the user Equipment and the gNB each comprises a processor coupled with a memory in which computer program instructions are stored, said instructions being configured to implement steps of the claims 1 to 9.