Method of ai / ml model matching signaling

Abstract

The present application describes methods of data-driven AI / ML model signaling for AI / ML based pairable model search signaling, where techniques for pre-configuring and signaling about searching model pairing of machine learning model operations in wireless mobile communication systems including base stations (e.g., gNB) and mobile stations (e.g., UE) are disclosed. The minimum model information set is provided to facilitate the search for pairable model(s) to match through a model pairing search process.

Description

[0001] TITLE

[0002] Method of AI / ML model matching signaling

[0003] TECHNNICAL FIELD

[0004] The present disclosure relates to AI / ML based pairable model search signaling, where techniques for pre-configuring and signaling about searching model pairing of machine learning model operations are presented.

[0005] BACKGROUND

[0006] In 3GPP (3rd 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 RAN (Technical Specification Group Radio Access Network) meeting #94e. The official title of AI / ML study item is “Study on AI / ML for NR Air Interface”, and currently RAN WG1 and WG2 are actively working on a specification. 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.

[0007] 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 the framework will be one key work scope. Regarding AI / ML framework, various aspects are under consideration for investigation and one key item is about lifecycle management of AI / ML models where multiple stages are included as mandatory for model training, model deployment, model inference, model monitoring, model updating etc.

[0008] Earlier, in 3GPP TR 37.817 for Release 17 titled “Study on enhancement for Data Collection for NR and EN-DC”, UE mobility was also considered as one of AI / ML use cases and one scenario for model training / inference is that both functions are located within a RAN node. Followingly, in Release 18 the new work item of “Artificial Intelligence (Al) / Machine Learning (ML) for NG-RAN” was initiated to specify data collection enhancements and signaling support within existing NG-RAN interfaces and architecture.

[0009] In 3GPP AI / ML discussions, there are several working assumptions including terminologies. For example, a two-sided (AI / ML) model is a paired AI / ML model over which joint inference is performed, where joint inference comprises AI / ML inference which is jointly performed across the UE and the network, i.e. , the first part of inference is firstly performed by a UE and then the remaining part is performed by a gNB, or vice versa. Proprietary-format models are ML models of vendor-Zdevice-specific proprietary format from 3GPP perspective and they are not mutually recognizable across vendors and hide model design information from other vendors when shared. Open-format models are ML models of specified format that are mutually recognizable across vendors and allow interoperability from 3GPP perspective and they are mutually recognizable between vendors and do not hide model design information from other vendors when shared.

[0010] For the above active standardization works, currently there is no specification defined for signaling methods or network (e.g., gNB, BS) / mobile station (e.g., UE) behaviors about supporting AI / ML model operation of model pairing between two nodes when multiple LCM (lifecycle management) operations are enabled on a device such as model training and inferencing and / or updating, etc., for one or more particular application scenarios with different sets of ML features / functionalities. When model identifications (e.g., model ID) are not pre-configured for model pairing, it is quite difficult to configure two-sided models and significant delay for configuration can occur to find the matched model pair(s).

[0011] US 2021 344469 A1 shows that a machine learning model is implemented in the UE and the UE estimates one or more features of a second band based on measurements it performs on a reference signal transmitted from a dedicated transmitter in a first band. US 2022 149 980 A1 shows amethod for dynamically selecting a link adaptation policy, where the method includes using channel quality information, additional information, and an ML model to select a link adaptation policy from a set of predefined link adaptation policies.

[0012] US 2022 150 727 A1 shows techniques for sharing machine learning models and an indication of transmission and reception points (TRPs) for which the machine learning models are applicable between wireless nodes such as UEs and BSs.

[0013] US 2023 153 408 A1 shows a method including the steps of creating, using a machine learning model being trained, calculating, and updating parameters of the machine learning model based on a calculated loss.

[0014] BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The disclosed invention will be further discussed in the following based on preferred embodiments presented in the attached drawings. However, the disclosed invention may be embodied in many different forms and should not be construed as limited to said preferred embodiments. Rather, said preferred embodiments are provided for thoroughness and completeness, and fully convey the scope of the invention to the skilled person. The following detailed description refers to the attached drawings, in which:

[0016] Figure 1 is an exemplary block diagram of a blind pairing search method for ML models;

[0017] Figure 2 is an exemplary block diagram of a minimum model information set table;

[0018] Figure 3 is a signaling flow of searching candidate pairable models for model matching;

[0019] Figure 4 is a flowchart of a procedure of blind pairing search for a network side; Figure 5 is a flowchart of a procedure of blind pairing search for a UE side;

[0020] Figure 6 is a flowchart of a procedure of determining combination of model pair(s); and

[0021] Figure 7 is a flowchart of a procedure of determining types of candidate pairable models.

[0022] DETAILED DESCRIPTION

[0023] The detailed description set forth below, with reference to the 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.

[0024] 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.

[0025] 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.

[0026] 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), another UE, etc.

[0027] 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 Ml, UE category M2, ProSe UE, V2V UE, V2X UE, etc. Additionally, terminologies such as base station / gNodeB and UE should be considered non-limiting and in particular do 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.

[0028] As will be appreciated by one skilled in the art, aspects of the embodiments may be embodied as a system, apparatus, method, or computer program product.

[0029] 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.

[0030] 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.

[0031] Furthermore, embodiments may take the form of a computer 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

[0032] 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.

[0033] 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 that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device.

[0034] 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”)).

[0035] 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.

[0036] According to a first aspect of the invention, a method of AI / ML model matching signaling between a first node and a second node in a wireless network to support an AI / ML model pairing procedure between a model pair consisting of a first AI / ML model of the first node and a second AI / ML model of the second node, comprises a step of receiving an ML configuration information which indicates a model pairing request matching the first AI / ML model. Further, the method comprises a step of identifying a finite list of candidate pairable models. Further, the method comprises a step of fenerating at least one minimum model information set for each of the candidate pairable models, each of the at least one minimum model information set containing descriptive information about its respective candidate pairable model. Further, the method comprises a step of receiving the at least one minimum model information set at the first node. Further, the method comprises a step of determining, using the at least one minimum model information set, the second AI / ML model as that candidate pairable model, which yields a best combination with the first AI / ML model. Advantageously, the determination of the second AI / ML model is based on a measurement of a performance on a target performance metric of the model pair.

[0037] Advantageously, the at least one minimum model information set contain information about model functionality, and / or model input / output, and / or model description, and / or model meta data.

[0038] Advantageously, the determination of the second AI / ML model utilizes a blind pairing search.

[0039] Advantageously, the at least one minimum model information set is transmitted from the second node to the first node via L1 , L2, or L3 signaling.

[0040] Advantageously, the at least one minimum model information set is indexed in a pre-configured format to support the generation of the finite list of candidate models.

[0041] Advantageously, the at least one minimum model information set holds multiple tables.

[0042] Advantageously, the candidate pairable models are classified as either a common reference model or a dedicated reference model, the method further comprising the steps:

[0043] • Receiving at least one pre-configured range regarding the at least one minimum model information set,

[0044] • Selecting the common reference model as the second AI / ML model if the at least one model information set is below the pre-configured range,

[0045] • Selecting the dedicated reference model as the second AI / ML model if the at least one minimum model information set is above the pre-configured range.

[0046] Advantageously, the number of candidate pairable models to be tested for model performance can be adjusted based on the implementation setting. Advantageously, a search failure is issued to enable an alternative AI / ML configuration or fallback operation if no model pair provides sufficient target performance metric.

[0047] Advantageously, the nodes can be a UE, a network, a base station, or a combination thereof, and the communication between the nodes can be performed via L1 , L2, or L3 signaling, or as sidelink-based communication.

[0048] According to a second aspect of the invention, an apparatus for AI / ML model matching signaling, comprising a processor coupled with a memory in which computer program instructions are stored, said instructions being configured to implement steps an above-described method.

[0049] According to a third aspect of the invention, a user equipment comprises the above-described apparatus, characterized in that the steps of receiving the ML configuration information, identifying the finite list of candidate pairable models, determining the at least one minimum model information set, and transmitting the at least one minimum model information set, are proceeded.

[0050] According to a fourth aspect of the invention, a base station (gNB) comprises the above-described apparatus, characterized in that the steps of transmitting the ML configuration information, receiving the at least one minimum model information set, and determining the second AI / ML model are proceeded.

[0051] According to a fifth aspect of the invention, a wireless communication system comprises the above-described base station (gNB) and the above-described user equipment, characterized in that the base station (gNB) comprises a processor coupled with a memory in which computer program instructions are stored, said instructions being configured to implement steps of an above-described method, wherein the user equipment (UE) comprises a processor coupled with a memory in which computer program instructions are stored, said instructions being configured to implement steps of an above-described method. Aspects of the embodiments are described below with reference to schematic flowchart diagrams and / or schematic block diagrams of methods, apparatuses, systems, and computer 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 are executed via the processor of the computer or other programmable data processing apparatus, create means for implementing the fimctions / acts specified in the flowchart diagrams and / or block diagrams.

[0052] 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.

[0053] 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.

[0054] 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). 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.

[0055] 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 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.

[0056] 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

[0057] The following explanation will provide the detailed description of the mechanism about data-driven AI / ML model signaling for ML model operation in wireless mobile communication systems including base stations (e.g., gNB) and mobile stations (e.g., UE).

[0058] An AI / ML lifecycle can be split into several stages such as data collection / pre-processing, model training, model testing / validation, model deployment / update, model monitoring, model switching / selection etc., where each stage is equally important to achieve target performance with any specific model(s). In applying an AI / ML model for any use case or application, one of the challenging issues is to manage the lifecycle of AI / ML model. This is mainly because of a data / model drift occurs during model deployment / inference which results in performance degradation of the AI / ML model. Fundamentally, statistical changes of datasets occur after the model is deployed and the model inference capability is impacted while using unseen data as input.

[0059] In a similar aspect, the statistical property of a dataset and the relationship between input and output for the trained model can change with drift occurrence. Then, model adaptation is required to support operations such as model switching, re-training, fallback, etc. Also model pairing is needed allowcooperation between two or more models located in different nodes such as gNB-UE or UE-UE. In model pairing, the important step is to find the best combination of paired models for different applications and / or functionalities using ML. Especially when model identifications (e.g., model ID) are not pre-configured for model pairing, it is quite difficult to configure model pairs. Significant delay for the associated configuration can occur to find the matched model pair whilesearching the candidate model pair(s).

[0060] In this method, the finite list of candidate pairable models to match with a UE model can be determined with minimum model information sets (e.g., model functionality, model input / output) to identify the best combination of model pair(s) through blind pairing search. For the blind pairing search method, minimum model information about the UE model is shared between two nodes such as network side and UE side. A finite number of candidate panable models are determined to be searched for model pairing. Each pairable models, i.e. the determined candidate pairable models and the UE model, are measured for model performance based on at least one target performance metric, such as a model accuracy measure. The minimum model information set is the information to be shared for supporting the search for a finite list of candidate models to be paired. The minimum model information sets contain model description information or meta data that can be shared. For example, when the UE has a model with no model ID and a minimum model information set is shared with the network side, the network side searches candidate pairable models using this shared information from the UE. Selection criteria for the blind pairing search is based on a pre-defined metric related to a model performance measure.

[0061] The key benefits include latency reduction for matching model pair(s) and improvement of searching model pairs when the explicit model identification is not available.

[0062] Figure 1 shows an exemplary block diagram of a blind pairing search method for ML models. In this exemplary block diagram, the scenario is that there are multiple candidate pairable models on the network side that can be potentially matched with a UE model through a blind search method to find the best combination of a network-UE model pair(s). There are one or more search steps to be taken before obtaining final candidate pairable models and each search step can proceed based on a minimum model information set(s) which is provided by the UE side. The minimum model information set is related to model attribute data such as model description, meta data, model functionality, model type / format, model input / out, etc. The UE model itself can be based on a proprietary-format or a hidden model design.

[0063] Depending on how much UE model information through the minimum model information set(s) can be shared with the network side, the number of search steps can be determined so that all potential pairable models can be found by narrowing down a finite list of candidate pairable models. After identifying candidate pairable models, a model performance based on target performance metric(s) (e.g., model accuracy) can be measured for each combination of model pairs as a selection criteria. The pairable model search is based on a pre-defined metric related to model performance. Although in this exemplary block diagram, multiple candidate pairable models at the network side have been identified to be possibly matched with a specific UE model, it is also applicable that there are multiple candidate pairable models determined at UE side that could be matched with a specific network side model. In addition, the same mechanism is applicable to sidelink based model pairing search between UEs. The minimum model information sets can be related to either UE side model or network side model or the combined UE / network side models in terms of model information that can help find the matched model pair(s).

[0064] Figure 2 shows an exemplary block diagram of minimum model information set table. In this exemplary table, the index and the associated format of minimum model information set can be pre-configured and this table contains the information to be shared with the other node side for helping search a finite list of candidate models to be paired. Minimum model information sets contain model attribute data such as model description, meta data, model functionality, model type / format, model input / out, etc. There can be multiple tables having different content of minimum information sets and / or number of indexes that can be pre-defined according to different ML applications and model functionalities, etc. The pre-configured table(s) of the minimum model information sets can then be sent to the other node side through L1 , L2 or L3 signaling.

[0065] Figure 3 shows a signaling flow of searching candidate pairable models for model matching. Based on ML configuration information provided by the network side, the minimum model information sets are determined to be shared with the network side so that a blind pairing search can be executed. The minimum model information set format related to the content of each set can be pre-configured so that each indexed set of minimum model information can be generated based on ML applications, applicable models, model functionalities, etc. After receiving the minimum model information sets from the UE, the blind pairing search is performed between the network and the UE to find the best combination of paired models among candidate pairable models.

[0066] Figure 4 shows a flowchart of a procedure of a blind pairing search for the network side. In this exemplary procedure, the network side pre-configures the format of the indexed minimum model information sets that are shared with the other node (e.g., a UE). The UE can then use this format to generate the available minimum model information set(s). The format of the indexed minimum model information sets can vary and there can be multiple formats for different ML applications, applicable models, model functionalities, etc. There might even be multiple formats for the same purpose to be shared with the UE (e.g., different sizes or versions of formats). Based on the received minimum model information set(s), network side executes a blind pairing search to find candidate pairable models using any reference model that can be determined by using the received minimum model information set(s).

[0067] Figure 5 shows a flowchart of a procedure of a blind pairing search for the UE side. After receiving the ML configuration information, the UE model is determined so that the associated minimum model information sets can be generated based on the pre-configured format with an index.

[0068] Figure 6 shows a flowchart of a procedure of determining a combination of model pair(s). After identifying candidate pairable models, the model performance is measured based on target performance metric(s) (e.g., model accuracy measure) for each combination of model pairs as a selection-criteria. The pairable model search is therefore based on pre-defined metrics related to model performance. Among the identified candidate pairable models, the number of models among the identified candidate pairable models to be tested for model performance can be adjusted based on the implementation setting. This model performance measurement step is also considered as pre-inferencing. After the model performance measurement with candidate pairable models, the best combination of the matched model pair(s) can be selected. If there is no matched model pair, a search failure might be envoked and an alternative ML configuration or fallback operation needs to be enabled.

[0069] Figure 7 shows a flowchart of a procedure of determining types of candidate pairable models. Two types of candidate pairable models can be defined such as a common reference model and a dedicated reference model. The common reference model is used to be paired with the other-sided model during or after the blind pairing search, wherein the shared minimum model information from the other-sided model is below a pre-configured range or threshold. The common reference model can be directly used among candidate pairable models or when searching for candidate pairable models. The selection of the reference model (either common or dedicated) might depend on how much minimum model information sets are available. The dedicated reference model is used to be paired with the other-sided model during or after the blind pairing search when the pre-configured range or threshold is not met. The shared minimum model information from the other-sided model is used, when the shared minimum model information is above the pre-configured range or threshold. Like the common reference model, the dedicated reference model can be directly used among candidate pairable models or when searching for candidate pairable models. The pre-configured range or threshold indicates how many sets of the minimum model information are shared between two.

[0070] Regarind network / UE side behaviors , both sides can be the node that performs the blind pairing search depending on different deployment scenarios, including sidelink-based communication. In the above-discussed figures, the format of the minimum model information sets is pre-configured on the network side wherein RRC(L3) signaling can be used to send the configuration information while other signaling types like L1 or L2 can also be considered. However, other node types such as UE or edge computing devices can perform the pre-configuration of the format of minimum model information sets if necessary.

[0071] This application provides fundamental mechanisms of interworking and data information flow in radio access network collaboration for AI / ML support, especially in searches for best combinations of UE models and associated pairable models in ML operation aspects.

[0072] Based on the proposed invention, gNB-UE behaviors for supporting AI / ML operations for wireless communication with joint ML operations can be greatly improved with the potential scenarios. Abbreviations:

[0073] Al Artificial intelligence BWP Bandwidth part CBG Code block group CLI Cross Link Interference CP Cyclic prefix CQI Channel quality indicator CPU CSI processing unit CRB Common resource block CRC Cyclic redundancy check CRI CSI-RS Resource Indicator CSI Channel state information CSI-RS Channel state information reference signal CSI-RSRP CSI reference signal received power CSI-RSRQ CSI reference signal received quality CSI-SINR CSI signal-to-noise and interference ratio CW Codeword DCI Downlink control information DL Downlink DM-RS Demodulation reference signals DRX Discontinuous Reception EPRE Energy per resource element IAB-MT Integrated Access and Backhaul - Mobile Terminal ID Identificator L1 -RSRP Layer 1 reference signal received power LI Layer Indicator LCM Life cycle management MCS Modulation and coding scheme ML Machine learning NW Network PDCCH Physical downlink control channel PDSCH Physical downlink shared channel PSS Primary Synchronisation signal PUCCH Physical uplink control channel QCL Quasi co-location PMI Precoding Matrix Indicator PRB Physical resource block PRG Precoding resource block group PRS Positioning reference signal PT-RS Phase-tracking reference signal RAN Radio Access Network RB Resource block RBG Resource block group Rl Rank Indicator RIV Resource indicator value RS Reference signal SCI Sidelink control information SLIV Start and length indicator value SR Scheduling Request SRS Sounding reference signal SS Synchronisation signal SSS Secondary Synchronisation signal SS-RSRP SS reference signal received power

[0074] SS-RSRQ SS reference signal received quality SS-SINR SS signal-to-noise and interference ratio TB Transport Block TCI Transmission Configuration Indicator TDM Time division multiplexing UE User equipment UL Uplink WG Work group

Claims

CLAIMS1 . A method of AI / ML model matching signaling between a first node and a second node in a wireless network to support an AI / ML model pairing procedure between a model pair consisting of a first AI / ML model of the first node and a second AI / ML model of the second node, the method comprising the steps:• Receiving an ML configuration information which indicates a model pairing request matching the first AI / ML model,• Identifying a finite list of candidate pairable models,• Generating at least one minimum model information set for each of the candidate pairable models, each of the at least one minimum model information set containing descriptive information about its respective candidate pairable model,• Receiving the at least one minimum model information set at the first node, and• Determining, using the at least one minimum model information set, the second AI / ML model as that candidate pairable model, which yields a best combination with the first AI / ML model.

2. The method according to claim 1 , characterized in that the determination of the second AI / ML model is based on a measurement of a performance on a target performance metric of the model pair.

3. The Method according to claim 1 or 2, characterized in that the at least one minimum model information set contain information about model functionality, and / or model input / output, and / or model description, and / or model meta data.

4. The method according to any of the previous claims, characterized in that the determination of the second AI / ML model utilizes a blind pairing search.

5. The method according to any of the previous claims, characterized in that the at least one minimum model information set is transmitted from the second node to the first node via L1 , L2, or L3 signaling.

6. The method according to any of the previous claims, characterized in that the at least one minimum model information set is indexed in a pre-configured format to support the generation of the finite list of candidate models.7.The method according to claim 5, characterized in that the at least one minimum model information set holds multiple tables.

8. The method according to any of the previous claims, characterized in that the candidate pairable models are classified as either a common reference model or a dedicated reference model, the method further comprising the steps:• Receiving at least one pre-configured range regarding the at least one minimum model information set,• Selecting the common reference model as the second AI / ML model if the at least one model information set is below the pre-configured range,• Selecting the dedicated reference model as the second AI / ML model if the at least one minimum model information set is above the pre-configured range.

9. The method according to claim 2, characterized in that the number of candidate pairable models to be tested for model performance can be adjusted based on the implementation setting.

10. The method according to claim 2, characterized in that a search failure is issued to enable an alternative AI / ML configuration or fallback operation if no model pair provides sufficient target performance metric.11 . The method according to any of the previous claims, characterized in that the nodes can be a UE, a network, a base station, or a combination thereof, and the communication between the nodes can be performed via L1 , L2, or L3 signaling, or as sidelink-based communication.

12. An apparatus for AI / ML model matching signaling, comprising 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 11 .

13. A user equipment, comprising the apparatus according to claim 12, characterized in that the steps of receiving the ML configuration information, identifying the finite list of candidate pairable models, determining the at least one minimum model information set, and transmitting the at least one minimum model information set, are proceeded.

14. A base station (gNB), comprising the apparatus according to claim 12, characterized in that the steps of transmitting the ML configuration information, receiving the at least one minimum model information set, and determining the second AI / ML model are proceeded.

15. A wireless communication system, comprising the base station (gNB) according to claim 14 and the user equipment according to claim 13, characterized in that the base station (gNB) comprises a processor coupled with a memory in which computer program instructions are stored, said instructions being configured to implement steps of claims 1 to 11 , wherein the user equipment (UE) 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 11 .

Images