AI / ML-related data collection

Abstract

Example embodiments of the present disclosure relate to AI / ML related data collection. A method includes receiving, at a first device from a second device, configuration information for monitoring of a group common physical downlink control channel (GC-PDCCH), the GC-PDCCH associated with artificial intelligence (AI) / machine learning (ML) related data collection; monitoring a GC-PDCCH associated with the AI / ML related data collection based on the configuration information; determining indication information for enabling or disabling of AI / ML related data collection from the detected GC-PDCCH in accordance with a determination that the GC-PDCCH is detected; and performing an action on the AI / ML related data collection process based on the indication information.

Description

Cross-reference to related applications

[0001] This application claims priority and interest in U.S. Provisional Application 63 / 516953, filed August 1, 2023, the contents of which are incorporated herein by reference in their entirety. Technical Field

[0002] Various exemplary embodiments of this disclosure are generally related to the telecommunications field, and more particularly to methods, apparatuses, devices, and computer-readable storage media for collecting data related to artificial intelligence / machine learning (AI / ML). Background Technology

[0003] In the telecommunications industry, artificial intelligence / machine learning (AI / ML) models have already been adopted in telecom systems to improve performance. For example, 3GPP Release 18 initiated research into AI / ML for the New Radio (NR) air interface. The goal is to explore the benefits of enhancing the air interface with features that can improve support for AI / ML-based algorithms to enhance performance and / or reduce complexity / overhead. Several use cases are considered capable of identifying public AI / ML frameworks, including the functional requirements of AI / ML architectures that can be used in subsequent projects. It also aims to identify areas where AI / ML can improve the performance of air interface functions. The impact on communication specifications will be assessed to improve the overall understanding of what will be needed to enable AI / ML technologies for the air interface. Summary of the Invention

[0004] In a first aspect of this disclosure, a first apparatus is provided. The first apparatus includes: at least one processor; and at least one memory storing instructions, which, when executed by the at least one processor, cause the first apparatus to perform at least: receiving configuration information from a second apparatus for monitoring a Group Common Physical Downlink Control Channel (GC-PDCCH), the GC-PDCCH being associated with artificial intelligence (AI) / machine learning (ML) related data collection; monitoring the GC-PDCCH associated with the AI / ML related data collection based on the configuration information; determining, based on the determination that the GC-PDCCH is detected, enabling or disabling indication information for the AI / ML related data collection from the detected GC-PDCCH; and performing an action on the AI / ML related data collection process based on the indication information.

[0005] In a second aspect of this disclosure, a second apparatus is provided. The second apparatus includes: at least one processor; and at least one memory storing instructions, which, when executed by the at least one processor, cause the second apparatus to at least: send configuration information associated with monitoring of a Group Common Physical Downlink Control Channel (GC-PDCCH) to a group of first apparatuses, the GC-PDCCH being associated with the collection of artificial intelligence (AI) / machine learning (ML) related datasets by the group of first apparatuses; determine whether the AI / ML related data collection should be enabled or disabled; and based on the determination of whether the AI / ML related data collection should be enabled or disabled, send indication information for enabling or disabling the AI / ML related data collection to the group of first apparatuses via the GC-PDCCH.

[0006] In a third aspect of this disclosure, a third apparatus is provided. The third apparatus includes: at least one processor; and at least one memory storing instructions, which, when executed by the at least one processor, cause the third apparatus to perform at least: in response to a data collection initiation request, allocating a temporary model identifier for associating data to be collected in an AI / ML-related data collection performed by a set of first apparatuses; sending a data collection request to a second apparatus, the data collection request including characteristics of the data to be collected and a temporary model identifier for associating the data to be collected in an AI / ML-related data collection performed by a set of first apparatuses; receiving from the second apparatus a data collection token assigned for the AI / ML-related data collection; and mapping the data collection token to the temporary model identifier.

[0007] In a fourth aspect of this disclosure, a method is provided. The method includes: receiving from a second device configuration information for monitoring a Group Common Physical Downlink Control Channel (GC-PDCCH), the GC-PDCCH being associated with artificial intelligence (AI) / machine learning (ML) related data collection; monitoring the GC-PDCCH associated with the AI / ML related data collection based on the configuration information; determining, based on the determination that the GC-PDCCH is detected, enabling or disabling the AI / ML related data collection from the detected GC-PDCCH; and performing an action on the AI / ML related data collection process based on the indication information.

[0008] In a fifth aspect of this disclosure, a method is provided. The method includes: sending configuration information associated with monitoring of a group common physical downlink control channel (GC-PDCCH) to a group of first devices, the GC-PDCCH being associated with artificial intelligence (AI) / machine learning (ML) related data collection performed by the group of first devices; determining whether the AI / ML related data collection should be enabled or disabled; and based on the determination of whether the AI / ML related data collection should be enabled or disabled, sending indication information via the GC-PDCCH for enabling or disabling the AI / ML related data collection.

[0009] In a sixth aspect of this disclosure, a method is provided. The method includes: in response to a data collection initiation request, assigning a temporary model identifier for associating data to be collected in an AI / ML-related data collection performed by a set of first devices; sending a data collection request to a second device, the data collection request including characteristics of the data to be collected and the temporary model identifier for associating the data to be collected in the AI / ML-related data collection performed by the set of first devices; receiving from the second device a data collection token assigned for the AI / ML-related data collection; and mapping the data collection token to the temporary model identifier.

[0010] In a seventh aspect of this disclosure, a first apparatus is provided. The first apparatus includes: components for receiving configuration information from a second apparatus for monitoring a Group Common Physical Downlink Control Channel (GC-PDCCH), the GC-PDCCH being associated with Artificial Intelligence (AI) / Machine Learning (ML) related data collection; components for monitoring the GC-PDCCH associated with AI / ML related data collection based on the configuration information; components for determining, based on the determination that the GC-PDCCH is detected, indication information for enabling or disabling AI / ML related data collection from the detected GC-PDCCH; and components for performing actions on the AI / ML related data collection process based on the indication information.

[0011] In an eighth aspect of this disclosure, a second apparatus is provided. The second apparatus includes: components for transmitting configuration information associated with monitoring of a group common physical downlink control channel (GC-PDCCH) to a group of first devices, the GC-PDCCH being associated with artificial intelligence (AI) / machine learning (ML) related data collection performed by the group of first devices; components for determining whether the AI / ML related data collection should be enabled or disabled; and components for transmitting indication information for enabling or disabling the AI / ML related data collection via the GC-PDCCH to the group of first devices based on the determination that the AI / ML related data collection should be enabled or disabled.

[0012] In a ninth aspect of this disclosure, a third apparatus is provided. The third apparatus includes: means for allocating a temporary model identifier for associating data to be collected in an AI / ML-related data collection performed by a set of first apparatuses in response to a data collection initiation request; means for sending a data collection request to a second apparatus, the data collection request including characteristics of the data to be collected and a temporary model identifier for associating the data to be collected in an AI / ML-related data collection performed by a set of first apparatuses; means for receiving a data collection token assigned to the AI / ML-related data collection from the second apparatus; and means for mapping the data collection token to the temporary model identifier.

[0013] In a tenth aspect of this disclosure, a computer-readable medium is provided. The computer-readable medium includes instructions stored thereon for causing a device to at least execute the method according to a fourth aspect.

[0014] In the eleventh aspect of this disclosure, a computer-readable medium is provided. The computer-readable medium includes instructions stored thereon for causing a device to at least execute the method according to the fifth aspect.

[0015] In a twelfth aspect of this disclosure, a computer-readable medium is provided. The computer-readable medium includes instructions stored thereon for causing a device to at least execute the method according to the sixth aspect.

[0016] It should be understood that the summary portion is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0017] Some exemplary embodiments will now be described with reference to the accompanying drawings, in which: Figure 1 An example communication environment in which example embodiments of the present disclosure may be implemented is shown; Figure 2 A flowchart of a signaling flow for network-assisted data collection according to some example embodiments of the present disclosure is shown; Figure 3A and Figure 3B A flowchart of a signaling flow for network-assisted data collection according to some example embodiments of the present disclosure is shown; Figure 4 A flowchart is shown illustrating a method implemented at a first device according to some exemplary embodiments of the present disclosure; Figure 5 A flowchart is shown illustrating a method implemented at a second device according to some example embodiments of the present disclosure; Figure 6A flowchart is shown illustrating a method implemented at a third device according to some example embodiments of the present disclosure; Figure 7 A simplified block diagram of a device suitable for implementing example embodiments of the present disclosure is shown; and Figure 8 A block diagram of an example computer-readable medium according to some example embodiments of the present disclosure is shown.

[0018] Throughout the accompanying drawings, the same or similar reference numerals denote the same or similar elements. Detailed Implementation

[0019] The principles of this disclosure will now be described with reference to some exemplary embodiments. It should be understood that these embodiments are described for illustrative purposes only and to assist those skilled in the art in understanding and implementing this disclosure, without imposing any limitation on the scope of this disclosure. The embodiments described herein can be implemented in various ways other than those described below.

[0020] In the following description and claims, unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains.

[0021] References to "an embodiment," "embodiment," "example embodiment," etc., in this disclosure indicate that the described embodiment may include a particular feature, structure, or characteristic, but not every embodiment needs to include that particular feature, structure, or characteristic. Furthermore, such phrases do not necessarily refer to the same embodiment. Additionally, when a particular feature, structure, or characteristic is described in connection with an embodiment, it is proposed that, whether explicitly described or not, its effect in conjunction with other embodiments on such feature, structure, or characteristic is within the knowledge of those skilled in the art.

[0022] It should be understood that although various elements may be described herein using prefixes such as “first,” “second,” etc., these elements should not be limited by these terms. These terms are used only to distinguish one element from another, and they do not restrict the order of the terms. For example, without departing from the scope of the example embodiments, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element. As used herein, the term “and / or” includes any and all combinations of one or more of the listed terms.

[0023] As used herein, “at least one of the following: a list of two or more elements” and “at least one of a list of two or more elements” and similar wording, where a list of two or more elements is connected by “and” or “or”, means at least any one of the elements, or at least any two or more of the elements, or at least all of the elements.

[0024] As used herein, unless explicitly stated otherwise, the action “in response to A” does not indicate that the action is performed immediately after “A” occurs and may include one or more intervention steps.

[0025] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments. As used herein, the singular forms “a,” “an,” and “the” are also intended to include the plural forms unless the context clearly indicates otherwise. It will be further understood that the terms “comprising,” “including,” “having,” “having,” “containing,” and / or “comprising” as used herein specify the presence of stated features, elements, and / or components, etc., but do not exclude the presence or addition of one or more other features, elements, components, and / or combinations thereof.

[0026] As used in this application, the term "circuit system" may refer to one or more of the following: (a) Hardware circuit implementation only (such as implementation in analog and / or digital circuits only), and (b) A combination of hardware circuitry and software, such as (if applicable): (i) A combination of (multiple) analog and / or digital hardware circuits with software / firmware, and (ii) Any part of the software (including the software of the multiple hardware processors, including the multiple digital signal processors), the software, and the memory (the multiple memory), which work together to enable a device (such as a mobile phone or a server) to perform various functions, and (c) (multiple) hardware circuits and / or (multiple) processors, such as (multiple) microprocessors or a portion thereof, which require software (e.g., firmware) for operation, but the software may not exist when it is not required for operation.

[0027] This definition of circuit system applies to all uses of the term in this application, including in any claim. As another example, as used in this application, the term circuit also covers only hardware circuitry or a processor (or processors) or a portion thereof and its accompanying software and / or firmware implementation. For example, and if applicable to a particular claim element, the term circuit system also covers baseband integrated circuits or processor integrated circuits for use in mobile devices or servers, cellular network devices or other computing or networking devices.

[0028] As used herein, the term "communication network" refers to a network that conforms to any suitable communication standard, such as New Radio (NR), Long Term Evolution (LTE), LTE-A Advanced (LTE-A), Wideband Code Division Multiple Access (WCDMA), High-Speed ​​Packet Access (HSPA), Narrowband Internet of Things (NB-IoT), etc. Furthermore, communication between terminal devices and network devices in the communication network can be performed according to any suitable generation of communication protocol, including but not limited to first-generation (1G), second-generation (2G), 2.5G, 2.75G, third-generation (3G), fourth-generation (4G), 4.5G, fifth-generation (5G), sixth-generation (6G) communication protocols and / or any other currently known or future-developed protocols. Embodiments of this disclosure can be applied to various communication systems. Given the rapid development of communications, there will naturally be future types of communication technologies and systems that can utilize them to implement this disclosure. This disclosure should not be construed as limiting its scope to the aforementioned systems.

[0029] As used herein, the term "network device" refers to a node in a communications network through which terminal devices access the network and receive services. Network devices can refer to base stations (BS) or access points (APs), such as Node B (NodeB or NB), evolved Node B (eNodeB or eNB), NR NB (also known as gNB), Remote Radio Unit (RRU), Radio Header End (RH), Remote Radio Header End (RRH), relay, Integrated Access and Backhaul (IAB) node, low-power node (such as femtoseconds, picoseconds), non-terrestrial network (NTN) or non-terrestrial network equipment (such as satellite network equipment, low Earth orbit (LEO) satellites, and geostationary Earth orbit (GEO) satellites), spacecraft network equipment, etc., depending on the terminology and technology applied. In some example embodiments, the Radio Access Network (RAN) split architecture includes a centralized unit (CU) and a distributed unit (DU) at the IAB donor node. An IAB node includes a mobile terminal (IAB-MT) portion that behaves similarly to a UE with respect to its parent node, and the DU portion of the IAB node behaves similarly to a base station with respect to the next-hop IAB node table.

[0030] The term "terminal device" refers to any terminal device capable of wireless communication. As an example and not a limitation, a terminal device may also be referred to as a communication device, user equipment (UE), subscriber station (SS), portable subscriber station, mobile station (MS), or access terminal (AT). Terminal devices can include, but are not limited to, mobile phones, cellular phones, smartphones, Voice over IP (VoIP) phones, wireless local loop phones, tablets, wearable terminal devices, personal digital assistants (PDAs), portable computers, desktop computers, image capture terminal devices (such as digital cameras), gaming terminal devices, music storage and playback devices, in-vehicle wireless terminal devices, wireless endpoints, mobile stations, laptop embedded devices (LEEs), laptop devices (LMEs), USB dongles, smart devices, wireless customer premises equipment (CPEs), Internet of Things (IoT) devices, watches or other wearable devices, head-mounted displays (HMDs), vehicles, drones, medical devices and applications (e.g., remote surgery), industrial devices and applications (e.g., robots and / or other wireless devices operating in the context of industrial and / or automated processing chains), consumer electronics devices, devices operating on commercial and / or industrial wireless networks, etc. The terminal device may also correspond to the mobile terminal (MT) portion of an IAB node (e.g., a relay node). In the following description, the terms "terminal device," "communication device," "terminal," "user equipment," and "UE" are used interchangeably.

[0031] As used herein, the terms “resource,” “transmission resource,” “resource block,” “physical resource block” (PRB), “uplink resource,” or “downlink resource” can refer to any resource used to perform communication (e.g., communication between a terminal device and a network device), such as resources in the time domain, resources in the frequency domain, resources in the spatial domain, resources in the code domain, or any other combination of time, frequency, spatial, and / or code domain resources used to implement communication. In the following, unless explicitly stated otherwise, resources in both the frequency and time domains will be used as examples of transmission resources used to describe some exemplary embodiments of this disclosure. Note that the exemplary embodiments of this disclosure are equally applicable to other resources in other domains.

[0032] As used herein, the term "model" refers to the relationship between inputs and outputs learned from training data, and thus can generate corresponding outputs for a given input after training. Model generation can be based on machine learning (ML) techniques. Machine learning techniques can also be referred to as artificial intelligence (AI) techniques. Typically, machine learning models can be built that receive input information and make predictions based on that input information. For example, a classification model can predict the category of input information within a predetermined set of categories. As used herein, "model" can also be referred to as a "machine learning model," "learning model," "machine learning network," or "learning network," and these terms are used interchangeably throughout this document.

[0033] To facilitate understanding of the terminology, some definitions for the AI / ML terminology list are provided below.

[0034] AI / ML Models: Data-driven algorithms that apply AI / ML techniques to generate a set of outputs based on a set of inputs.

[0035] AI / ML Model Delivery: The general term refers to delivering an AI / ML model from one entity to another in any way. Note: An entity can refer to a network node / function (e.g., gNB, Location Management Function (LMF)), UE, proprietary server, etc.

[0036] AI / ML model inference: The process of using a trained AI / ML model to produce a set of outputs based on a set of inputs.

[0037] AI / ML Model Testing: A sub-process of training used to evaluate the performance of the final AI / ML model using a different dataset than that used for model training and validation. Unlike AI / ML model validation, testing does not assume subsequent model tuning.

[0038] AI / ML model training: The process of training an AI / ML model in a data-driven manner [by learning input / output relationships] and obtaining a trained AI / ML model for inference.

[0039] AI / ML model delivery: AI / ML models are delivered over the air interface in a manner opaque to 3GPP signaling, with parameters of the model structure known at the receiving end or a new model having parameters. Delivery may contain a complete model or a partial model.

[0040] AI / ML Model Validation: A sub-process of training used to evaluate the quality of AI / ML models using a different dataset than the dataset used for model training. This helps in selecting model parameters that generalize to datasets other than those used for model training.

[0041] Data collection: The process by which network nodes, management entities, or user units collect data for the purpose of AI / ML model training, data analysis, and inference.

[0042] Joint learning / joint training: A machine learning technique that trains AI / ML models across multiple decentralized edge nodes (e.g., UE, gNB), with each decentralized edge node performing local model training using local data samples. This technique requires multiple interactions between models but does not require exchanging local data samples.

[0043] Function Identification: The process / method for identifying AI / ML functions used for mutual understanding between the network and the UE. Note: Information about AI / ML functions can be shared during function identification. AI / ML function residency depends on specific use cases and sub-use cases.

[0044] Model activation: Enables AI / ML models for specific functions.

[0045] Model deactivation: Deactivate AI / ML models for specific functions.

[0046] Model download: Transferring the model from the network to the UE.

[0047] Model Identification: The process / method for identifying the AI / ML model used for mutual understanding between the network (NW) and the UE. Note: The process / method for model identification may or may not be applicable. Note: Information about the AI / ML model may be shared during model identification.

[0048] Model monitoring: The process of monitoring the inference performance of AI / ML models.

[0049] Model parameter update: The process of updating the model parameters.

[0050] Model selection: The process of selecting the AI / ML model to be activated from among multiple models for the same AI / ML enabling features. Note: Model selection may or may not be performed simultaneously with model activation.

[0051] Model switching: Deactivate the currently active AI / ML model and activate different AI / ML models for specific functions.

[0052] Model update: The process of updating the model parameters and / or model structure.

[0053] Model upload: The transfer of the model from the UE to the network.

[0054] Network-side (AI / ML) models: AI / ML models, where inference is performed entirely at the network level.

[0055] Offline field data: Data collected from the field and used for offline training of AI / ML models.

[0056] Offline training: The AI / ML training process in which a model is trained based on a collected dataset, and where the trained model is later used or delivered for inference. Note: This definition is for guidance only. There may be cases that do not perfectly fit this definition but can still be classified as offline training by generally accepted conventions.

[0057] Online field data: Data collected from the field and used for online training of AI / ML models.

[0058] Online training: The AI / ML training process in which the model used for inference is trained (usually continuously) as new training samples arrive (near) real-time. Note: The concepts of (near) real-time and non-real-time are context-dependent and relative to the inference timescale. Note: This definition is for guidance only. There may be cases that do not perfectly fit this definition but can still be classified as online training by generally accepted conventions. Note: Fine-tuning / retraining can be done via online or offline training. (This note can be removed when we define the term fine-tuning.) Reinforcement learning (RL): The process of training an AI / ML model from inputs (also called states) and feedback signals (also called rewards) generated by the model's outputs (also called actions) in the environment in which the model interacts with it.

[0059] Semi-supervised learning: The process of training a model using a mixture of labeled and unlabeled data.

[0060] Supervised learning: The process of training a model from inputs and their corresponding labels.

[0061] Two-sided (AI / ML) model: A pair of AI / ML models on which joint inference is performed, wherein joint inference includes AI / ML inference jointly performed across the UE and the network, i.e., the first part of the inference is first performed by the UE, and then the remaining part is performed by the gNB, and vice versa.

[0062] UE-side (AI / ML) model: AI / ML model, whose inference is performed entirely at the UE.

[0063] Unsupervised learning: The process of training a model without labeled data.

[0064] Raw format models: From a 3GPP perspective, these are vendor / device-specific proprietary ML models. They cannot be mutually recognized across vendors and hide model design information from other vendors when shared. Note: The example is a device-specific binary executable format.

[0065] Open format models: From a 3GPP perspective, these are specified format ML models that are mutually identifiable across vendors and allow interoperability. They are mutually identifiable between vendors and do not hide model design information from other vendors when shared.

[0066] Figure 1 An example communication environment 100 in which exemplary embodiments of the present disclosure can be implemented is shown. It should be understood that the elements shown in the communication system 100 are intended to represent the main functions provided within the system. Based on this, Figure 1 The boxes shown refer to specific elements in a communication network that provide these main functions. However, some or all of the main functions represented can be implemented using other network elements. Furthermore, it should be understood that not all functions of a communication network are represented in the boxes. Figure 1 The text is depicted in the middle. Instead, it illustrates the functionality of illustrative embodiments for ease of explanation. Furthermore, Figure 1 The number of elements shown is for illustrative purposes only, and any number of elements may be present.

[0067] As shown, the communication environment 100 includes multiple communication devices, including one or more first devices 110-1, 110-2, ..., 110-N (collectively or individually referred to as first devices 110) and one or more second devices 120. Figure 1 In the example, the second device 120 may include a network device, and the first device 110 may include a terminal device. The service area of ​​the second device 120 may be referred to as a cell. The first device 110 and the second device 120 may operate in a radio access network (RAN). Although two terminal devices (e.g., the first device) are shown, there may be more or fewer terminal devices in the service area of ​​the network device, and there may also be more network devices serving the terminal devices in the communication environment 100.

[0068] In the following description, for illustrative purposes, some exemplary embodiments are described in which the first device 110 operates as a terminal device and the second device 120 operates as a network device. However, in some exemplary embodiments, the operations described in connection with the terminal device can be implemented at the network device or other devices, and the operations described in connection with the network device can be implemented at the terminal device or other devices.

[0069] In some example embodiments, if the first device 110 is a terminal device and the second device 120 is a network device, the link from the second device 120 to the first device 110 is referred to as a downlink (DL), and the link from the first device 110 to the second device 120 is referred to as an uplink (UL). In the DL, the second device 120 is a transmitting (TX) device (or transmitter), and the first device 110 is a receiving (RX) device (or receiver). In the UL, the first device 110 is a TX device (or transmitter), and the second device 120 is an RX device (or receiver).

[0070] Communication in communication environment 100 can be implemented according to any suitable communication protocol(s), including but not limited to cellular communication protocols such as first-generation (1G), second-generation (2G), third-generation (3G), fourth-generation (4G), fifth-generation (5G), and sixth-generation (6G), wireless local area network communication protocols such as IEEE 802.11, and / or any other currently known or future-developed protocols. Furthermore, communication can utilize any suitable wireless communication technology, including but not limited to: Code Division Multiple Access (CDMA), Frequency Division Multiple Access (FDMA), Time Division Multiple Access (TDMA), Frequency Division Duplex (FDD), Time Division Duplex (TDD), Multiple-Input Multiple-Output (MIMO), Orthogonal Frequency Division Multiple Access (OFDM), Discrete Fourier Transform Extended OFDM (DFT-s-OFDM), and / or any other currently known or future-developed technologies.

[0071] In some example embodiments, one or more AI / ML models 105-1, 105-2, ..., 105-M (collectively or individually referred to as AI / ML model 105). AI / ML model 105 may sometimes be simply referred to as an AI model or an ML model. Different AI / ML models 105 may be configured to implement the same different algorithms in communication environment 100. Inference, testing, training, and / or verification of AI / ML model 105 may be performed at one or more of the first device 110, the second device 120, and / or other entities such as the third device 130. AI / ML model 105 may be delivered from one entity to another in any manner. AI / ML model 105 may be delivered over the air interface in a manner opaque to 3GPP signaling, whether it is a model structure known at the receiving end or a new model with parameters. Delivery may contain a complete model or a partial model.

[0072] The third device 130 may be an entity that manages the data collection of at least one or more AI / ML models 105. The third device 130 may be a network node in a radio access network (RAN) or (core network), or it may be an external entity. In some example embodiments, the third device 130 may be able to communicate with the second device 120 (e.g., a network device).

[0073] For the purposes of AI / ML model training, data analysis, and inference, a data collection process is typically required to collect data by network devices, management entities, or end devices. The collection of data for AI / ML models may be referred to herein as “AI / ML-related data collection.” For a specific AI / ML model(s), AI / ML-related data collection may be performed at one or more of the following entities: first device 110, second device 120, and / or other entities such as third device 130.

[0074] Within the general framework for AI / ML (models and functions), two distinct ML-related identifier types (function identifier, model identifier) ​​have been introduced for later discussion to distinguish between AI / ML models and the functions supported by AI / ML models. It is assumed that the model identifier uses "model ID" during the identifier process, and that the function identifier uses "function" during the identifier process. Definitions for the terms "model identifier" and "function identifier" are provided in Table 1 below. Note that whether and how functions are indicated will be discussed separately.

[0075] Table 1

[0076] Some further protocols related to functional-LCM (Lifecycle Management) and model ID-based LCM are shown in Tables 2 and 3 below.

[0077] Table 2

[0078] Table 3

[0079] Further explanation regarding model identification is provided in Table 4 below.

[0080] Table 4

[0081] Based on the above protocol, the following methods can be used to support AI / ML for the UE side or UE portion of the dual-side model identifier. Mode 1: Function-based Send "Conditions" in the UE capability report. NW uses "conditions" to configure functions for the UE. LCM can be performed within the configured features; Mode 2: UE-assisted functions Send "Conditions" in the UE capability report. NW uses "conditions" to configure functions for the UE. The UE can report the applicable functions among the configured functions (to resolve "additional conditions" and / or internal conditions). LCM can be executed within applicable functions; Mode 3: Function- and Model-Based The UE model is identified via an offline method (based on the gNB knowing the model ID, associated conditions, and additional conditions). Send the "conditions" (including the model ID) in the UE capability report. NW uses "conditions" (including the model ID) to configure functionality for the UE. LCM can be executed within the configured functions / models; Mode 4: UE-assisted functions & models The UE model is identified via an offline method (based on the gNB knowing the model ID, associated conditions, and additional conditions). Send the "conditions" (including the model ID) in the UE capability report. NW uses "conditions" (including the model ID) to configure functionality for the UE. The UE can report the applicable functions / models among the configured functions (to address "additional conditions" and / or internal conditions). LCM can be executed within the configured functions / models.

[0082] Regarding data collection, several protocols already exist related to data collection for AI / ML operations, and some of these protocols, applicable to all sub-use cases, are listed in Table 5.

[0083] Table 5

[0084] Specifically, for beam prediction using AI / ML, the following protocols in Table 6 were considered.

[0085] Table 6

[0086] Based on the above discussion, the network can trigger or request data collection from the UE, and can utilize various forms of auxiliary information for data collection.

[0087] Data collection is considered necessary for a variety of purposes, such as model training, feature / model validation, performance monitoring (of features / models), and model updates, and even for inference (e.g., beam prediction in the time domain). As listed above, several protocols exist regarding various enhancements related to data collection on the UE side. The inventors have identified the following challenges in data collection that have not yet been fully addressed: As mentioned above, in AI / ML use cases within the air interface, the AI / ML 3GPP framework is considered to be developed based on function and / or model identification, and the AI / ML LCM aspect can also be handled by function-LCM and model ID-LCM. However, the data collection aspects are developed in parallel and are not fully integrated with the function or model identification framework.

[0088] Data collection for the UE-side model can be used for model training and updates, where the UE may need to collect data over time using either a traditional measurement configuration or an enhanced measurement configuration (e.g., enhanced RS transmissions for data collection at the UE). The use of traditional measurements is the most attractive option because it avoids the air interface overhead of sending dedicated RS for data collection. However, the UE can still seek some ancillary information associated with data collection to make the data collection process smooth and efficient. 3GPP has not yet defined details regarding how the gNB provides ancillary information and what it provides as ancillary information.

[0089] For data collection and subsequent model development (training / updating the model based on the collected data), UE providers can use multiple UEs within a cell / site / area and collect data in parallel, uploading the collected data to the UE server. If data collection is performed incorrectly, the model updated / trained based on such data may not provide good performance for air interface use cases. This incorrect data collection may occur due to changes at the gNB that are not visible to the UE side. Consider the example of beam prediction; when the UE actively collects data, the NW may change the beam codebook, transmit power, and other NW considerations when transmitting RS within the cell / site / area.

[0090] According to some example embodiments of this disclosure, a solution for AI / ML related data collection is provided. A second device (e.g., a network device) configures a group of first devices to monitor a Group Common Physical Downlink Control Channel (GC-PDCCH), which is associated with artificial intelligence (AI) / machine learning (ML) related data collection performed by the group of first devices (e.g., terminal devices). The first devices receive configuration information for monitoring the GC-PDCCH and monitor the GC-PDCCH according to the configuration information. The second device determines whether AI / ML related data collection should be enabled or disabled, and based on the disadvantage of whether AI / ML related data collection should be enabled or disabled, sends indication information for enabling or disabling AI / ML related data collection to the group of first devices via the GC-PDCCH. Upon detecting the GC-PDCCH, the first devices can derive the indication information and perform actions on the AI / ML related data collection process based on the indication information.

[0091] This solution allows a second device to effectively (via GC-PDCCH) notify a group of first devices when network conditions or the environment suddenly change in a way that affects data collection on the first device side. This ensures that data collection is meaningful, i.e., that the collected data can be used in a meaningful way (e.g., if the number of beams being measured or their power changes, this needs to be taken into account in the measurement results of the collected data).

[0092] The exemplary embodiments of this disclosure will now be described in detail with reference to the accompanying drawings.

[0093] Figure 2 A flowchart of a signaling flow 200 for network-assisted data collection according to some example embodiments of the present disclosure is shown. For discussion purposes, reference is made to... Figure 1 To describe signaling flow 200. For example... Figure 2As shown, signaling flow 200 involves a group of first devices 110 (including one or more first devices) and second devices 120. In some example embodiments, one or more first devices 110 may be one or more terminal devices, and the second device 120 may be a network device (e.g., gNB, gNB CU, or gNB DU, etc.).

[0094] In signaling flow 200, it is assumed that a group of first devices 110 supports AI / ML-related data collection, for example, for developing AI / ML-enabled features. A second device 120 sends (205) configuration information associated with GC-PDCCH monitoring to the group of first devices 110, whereby the GC-PDCCH is associated with AI / ML-related data collection performed by the group of first devices. The second device 120 uses signaling to define the configuration information for detecting the GC-PDCCH in response to information related to AI / ML-related data collection. The configuration information enables the group of first devices 110 to monitor the GC-PDCCH.

[0095] In some example embodiments, configuration information may indicate at least one monitoring instruction for a GC-PDCCH associated with AI / ML-related data collection (e.g., a GC-PDCCH associated with UE-side data collection).

[0096] In some example embodiments, monitoring instructions may indicate or include information specific to the Data Collection Dedicated Radio Network Temporary Identifier (RNTI), information relating to the interpretation of downlink control information (DCI) carried in the GC-PDCCH, configuration of at least one associated AI / ML function for AI / ML-related data collection (e.g., configuration of associated ML function for data collection), and / or resource configuration for GC-PDCCH monitoring, and / or other PDCCH monitoring-related instructions. Resource configuration for GC-PDCCH monitoring may include, but is not limited to, search space, monitoring instances, monitoring resources (e.g., control resource sets, CORESET), etc.

[0097] In some example embodiments, the information associated with the interpretation of the DCI carried in the GC-PDCCH indicates at least one of the following: the starting position of the information within the DCI carried in the GG-PDCCH, an identifier of a predefined DCI format (e.g., the DCT format X_Y, which will be discussed below), or a list of identifiers of AI / ML models associated with AI / ML-related data collection to be enabled or disabled.

[0098] In some example embodiments, for communication specifications in which PDCCH monitoring can be captured (e.g., in 3GPP TS 38.212), solutions related to the configuration information of GC-PDCCH monitoring associated with AI / ML-related data collection can be specified as follows: For UEs configured with AI / ML data collection, the following can be provided for detecting DCI format X_Y in PDCCH reception. For DC-RNTI of DCI format X_Y via dc-RNTI, The number of search space sets for DCI format X_Y is detected on the active DL BWP of PCell or SpCell using dci-FormatX-X to monitor PDCCH according to the public search space as described below. For the payload size of DCI format X_Y via sizeDCI-XY; The position of the DCI format X_Y is initialized / stopped via data collection from DC-PositionDCI-XY. A value of '0' for the data collection initialization / stop indicator bit indicates the start of the data collection process; A value of '1' for the data collection initialization / stop indicator bit indicates that the data collection process has stopped; Bitmap, when configuring higher-layer parameters of the UE model-ID-for- When using DataCollection, where The bitmap position immediately follows the data collection initialization / stop indicator bit; The bitmap size is equal to the size of the bitmap. model-ID-for- The number of model IDs configured in the DataCollection, where each bit of the bitmap corresponds to a model ID; The '0' value of a bit used in a bitmap indicates that... model-ID-for- The model ID provided by DataCollection is valid for the data collection process; The '1' value of a bit in the bitmap indicates that... model-ID-for- The model ID of a DataCollection is invalid for the data collection process.

[0099] One or more first devices 110 receive (210) configuration information and monitor (215) GC-PDCCH associated with AI / ML related data collection based on the configuration information.

[0100] The second device 120 determines (220) whether AI / ML-related data collection should be enabled or disabled. In some cases, a vendor may use multiple first devices 110 (e.g., terminal devices) within a cell, site, or area, and collect data in parallel and upload the collected data to a server. Data collection is imperfect if the terminal devices (multiple) and network devices are not fully aware of the resources used for data collection, and therefore, for example, some network resource allocation characteristics may be poorly learned. Such erroneous data collection may occur due to changes at the network device level that are not visible to the terminal side. Consider the example of beam prediction, when a terminal device actively collects data, the network device may change the beam codebook, power variations, and other network-side considerations when transmitting a reference signal (RS) within the cell / site / area.

[0101] Based on the need to enable or disable AI / ML related data collection, the second device 120 sends (225) an instruction message via GC-PDCCH to a group of first devices 110 to enable or disable AI / ML related data collection. Through GC-PDCCH, for example, when network conditions change and therefore affect data collection, the second device 120 can effectively assist multiple first devices 110 in enabling or disabling AI / ML related data collection. The transmission of GC-PDCCH via the second device 120 can also be based on configuration information.

[0102] If one or more first devices 110 detect GC-PDCCH, each of the first devices 110 determines (230) an indication for enabling or disabling AI / ML related data collection from the detected GC-PDCCH. The first device 110 performs (235) an action on the AI / ML related data collection process based on the indication information.

[0103] On the first device side, the first device 110 monitors the GC-PDCCH associated with AI / ML related data collection, as instructed in the configuration information. After decoding the GC-PDCCH, the first device 110 can determine the indication information from the GC-PDCCH.

[0104] In some example embodiments, the indication information may include a first indicator indicating the enabling or disabling of AI / ML related data collection. In some example embodiments, the indication information may also include a second indicator indicating at least one identifier (ID) of at least one AI / ML model that contributes to / influences data collection. The ID of the AI / ML model (also referred to as the model ID) may be identified by the second device 120.

[0105] In some example embodiments, indications for enabling or disabling AI / ML-related data collection can be carried in the GC-PDCCH according to a predefined DCI format. The first indicator can be defined as a data collection initialization / stop indicator (e.g., a 1-bit field) having a value indicating enable and another value indicating disable.

[0106] In some example embodiments, the second indicator may include a bitmap, where bits in the bitmap correspond to the IDs of the AI / ML models identified by the second device 120. The number of bits in the bitmap may be determined based on the number of different model IDs. For example, the DCI carrying indicator information may include a field defined as “auxiliary information indicator,” which may be 0 bits if no higher-level parameter model-ID-for-DataCollection is configured; otherwise, a 1, 2, 3, 4, or L-bit bitmap is determined based on the number of different model IDs provided by the higher-level parameter model-ID-for-DataCollection, where each bit corresponds to one of the multiple model IDs configured by the higher-level parameter model-ID-for-DataCollection, wherein the most significant bit (MSB) to the least significant bit (LSB) of the bitmap corresponds to the first to the last configured model ID in ascending order of model ID.

[0107] In some example embodiments, for communication specifications in which the DCI format can be defined (e.g., in 3GPP TS 38.212), the solutions related to the DCI format can be specified as follows: The DCI format X_Y is used to notify changes in data collection for one or more UEs.

[0108] The following information is transmitted via DCI format X_Y with CRC scrambled by DC-RNTI: Block number 1, block number 2, ..., block number N; The starting position of the block is provided by a parameter from a higher layer for the UE configured with that block. DC-PositionDCI- XY Sure.

[0109] If the UE is configured with higher-level parameters dc-RNTI and dci-FormatX-Y Then, the higher layer configures a block for the UE, defining the following fields for that block: Data collection initialization / stop indication -1 bit Auxiliary information indicator -0 bit, if no higher-level parameters are configured. model-ID-for- DataCollection; otherwise, it depends on the higher-level parameters. model-ID-for- The number of different model IDs provided by the DataCollection determines a 1, 2, 3, 4, or L-bit bitmap, where each bit corresponds to a parameter from a higher level. model-ID-for- DataCollection n is one of the (multiple) model IDs configured, where the MSB to LSB of the bitmap corresponds to the first to the last configured model ID in ascending order of model ID.

[0110] The size of X_Y in DCI format is determined by higher-level parameters. sizeDCI-XY …instruct...

[0111] In some example embodiments, the RRC parameters used may be described, for example, in 3GPP TS 38.331 as follows: DC-PositionDCI-X-Y The starting position of the data collection initialization / stop indication in the DCI format XY (see TS 38.213…).

[0112] dci-FormatX-Y If configured, the UE monitors the DCI format X_Y according to TS 38.213 (clause…).

[0113] model-ID-for-DataCollection This field contains the model ID associated with data collection enabled / disabled via DCI format XY.

[0114] model-ID The model IDs, originating from the network and identified via the UE capability identifier, are configured for data collection at the UE. Configured in model-ID-for-DataCollection Inside.

[0115] In some example embodiments, if a first indicator decoded from the GC-PDCCH indicates the enabling of AI / ML related data collection, the first device 110 can initiate or resume an AI / ML related data collection process for at least one AI / ML model indicated by a second indicator in the GC-PDCCH. For example, if the AI / ML related data collection process for at least one AI / ML model was previously suspended, the first device 110 can resume the AI / ML related data collection process. Otherwise, the first device 110 can initiate a new AI / ML related data collection process to begin collecting data. In some example embodiments, if the first indicator indicates the disabling of AI / ML related data collection, the first device 110 can suspend or terminate the AI / ML related data collection process for at least one AI / ML model indicated by the second indicator.

[0116] In some example embodiments, if the AI / ML related data collection process is paused or terminated, the second device 120 may send resource authorization to a group of first devices 110. The first devices 110 may use the resource authorization to send data collected after the pause or termination of the AI / ML related data collection process to the second device 120.

[0117] In some example embodiments, the first device 110 may send capability signaling to indicate its capabilities associated with AI / ML. In some example embodiments, the AI / ML model may be identified by the network prior to capability signaling (e.g., UE capability signaling), which indicates AI / ML-enabled features or any other related features (e.g., data collection) associated with the AI / ML. The first device 110 may receive an assignment of at least one ID of at least one AI / ML model identified by the second device 120.

[0118] The AI / ML model ID is assigned to identify at least one of the following: a first condition reported in the capability signaling of the first device 110 for the AI / ML-enabled feature, or a second condition not reported in the capability signaling. In some example embodiments, the second device 120 (e.g., a network device) and (multiple) UE vendors may consider an offline method such that the model ID is assigned to identify the condition (which is reported in the UE capability signaling for the AI / ML-enabled feature) and to identify additional conditions (which may define site-specific, cell-specific, scenario-specific, or other relevant aspects not directly defined by the UE capability signaling). In some example embodiments, the second device 120 (e.g., a network device) and the first device 110 (e.g., a UE) may consider an online model identification process such that the model ID is assigned to identify the condition (which is reported in the UE capability signaling for the AI / ML-enabled feature) and to identify additional conditions (which may define site-specific, cell-specific, scenario-specific, or other relevant aspects not directly defined by the UE capability signaling). In some example embodiments, the first condition may refer to a condition, and the second condition may refer to an additional condition.

[0119] In some example embodiments, the AI / ML-enabled features and AI / ML-related data collection can be defined separately in the capability signaling to allow more first devices to allow data collection even if they do not fully support the AI / ML-enabled features(s). Specifically, the capability signaling of the first device 110 may indicate at least one of the following: an indication of whether the AI / ML-enabled features are supported by the first device 110, and / or an indication of whether AI / ML-related data collection is supported by the first device 110.

[0120] In some example embodiments, the second device 120 may anticipate that when the first device 110 reports an indication in capability signaling indicating its support for the AI / ML-related data collection process used for model development, it may indicate a model ID associated with the AI / ML-related data collection. Therefore, the capability signaling may additionally or alternatively indicate at least one ID of at least one AI / ML model associated with the AI / ML-related data collection supported by the first device 110. In some cases, some of the first devices 110 supporting data collection may not support the AI / ML-enabled features(s) and may only act as data collection entities within the network. In some cases, once data is collected at the first device 110, it can be assumed that a UE vendor-specific dataset is delivered to the UE server. As mentioned, a first device 110 that does not support AI / ML-enabled features may still indicate certain model IDs or other indicators associated with conditions and additional conditions. Furthermore, as indicated above, AI / ML-related data collection features may be defined as features independent of other AI / ML-enabled features and may also be associated with more than one model ID in the capability signaling of the first device 110.

[0121] Through the above example embodiments, the second device 120 can control the collection of AI / ML related data at the first device 110 by configuring and indicating instruction information, so as to enable or disable the AI / ML related data collection process by considering multiple first devices 110 simultaneously in a dynamic and efficient manner.

[0122] In some example embodiments, the second device 120 can format data collection requests to one or more first devices 110. The one or more first devices 110 can be selected by an algorithm within the second device 120. For example, the second device 120 can group first devices 110 that report the same model ID and have common capabilities regarding AI / ML-related data collection (e.g., the ability to collect beam level reference signal received power (RSRP) samples for up to 64 Synchronous SignAI / PBCH block (SSB) beam indices). The second device 120 can sort the configuration based on one or more sets of conditions reflecting possible changes in network configuration. For example, the second device 120 can switch from 16 SSB beams to 64 SSB beams during peak hours and switch back to 16 SSB beams due to energy savings during off-peak hours. Since this affects the AI / ML-related data collection process, it is important to track this on the first device 110 side when requesting AI / ML-related data collection. This is referred to as "network configuration conditions" or "condition set," which may have some impact on the collection of AI / ML-related data on the first device 110 side. It is expected that the (multiple) first devices 110 will achieve this in a manner that does not mix data samples across different "network configuration conditions."

[0123] In some example embodiments, the second device 120 may determine whether AI / ML-related data collection should be enabled or disabled based on a match between at least one condition of at least one AI / ML model associated with AI / ML-related data collection and network configuration conditions for transmitting signals collected by a set of first devices for such data. The second device 120 may match network configuration conditions with conditions(s) of at least one AI / ML model, such as conditions reported in the capability signaling of the first device 110 for enterprise AI / ML characteristics, as well as additional conditions defining site-specific, cell-specific, scenario-specific, or other relevant aspects that are not directly defined by the capability signaling of the first devices(s) 110.

[0124] In some example embodiments, if the network configuration conditions at the second device 120 remain under the first network configuration conditions and the first configuration conditions match at least one condition of at least one AI / ML model, then the second device 120 may initiate a set of AI / ML-related data collection procedures at the first device 110, wherein the first network configuration conditions may correspond to conditions and / or additional conditions associated with the model(s) ID(s) initiated by the AI / ML-related data collection procedures. The second device 120 may send first indication information via GC-PDCCH to the set of first devices 110 for enabling AI / ML-related data collection procedures for at least one AI / ML model. For example, a first indicator in the first indication information may instruct the set of first devices 110 to initiate AI / ML-related data collection procedures, and a second indicator in the first indication information may indicate the model(s) ID(s) of at least one AI / ML model.

[0125] In some example embodiments, if the second device 120 changes from a first network configuration condition to a second network configuration condition and the second network configuration condition does not match at least one condition of at least one AI / ML model, the second device 120 may suspend or terminate the AI / ML related data collection process on the first device 110 side, wherein the second network configuration condition may not involve additional conditions associated with the model(s) ID(s) of the ongoing AI / ML related data collection process. Specifically, the second device 120 may send a second instruction message via GC-PDCCH to a group of first devices 110 for disabling the AI / ML related data collection process for at least one AI / ML model.

[0126] In some example embodiments, if the network configuration conditions at the second device 120 switch back to the first network configuration conditions, the second device 120 can restart or initiate the data collection process at the first device 110.

[0127] In some example embodiments, the first device 110 may classify the collected data according to instructions received by the second device 120 and use it offline for model development.

[0128] Figure 3A and Figure 3B A flowchart of a signaling flow 300 for network-assisted data collection according to some example embodiments of the present disclosure is shown. For example... Figure 3A and Figure 3B As shown, signaling process 300 involves Figure 1A set of first devices (which may include one or more first devices 110), a second device 120, and a third device 130. In some example embodiments, one or more first devices 110 may be one or more terminal devices, the second device 120 may be a network device (e.g., gNB, gNBCU, or gNB DU, etc.), and the third device 130 may be a data collection entity (DCE) configured to manage data collection for AI / ML models and / or AI / ML functions. In some example embodiments, although shown as a separate entity, the third device 130 may be included in or implemented as either the first device 110 or the second device 120.

[0129] The third device 130 receives a data collection initiation request (305A, 305B, or 305C). In some example embodiments, the data collection initiation request may be sent from an external entity (e.g., an application service that expects to initiate a data collection request from the third device 130). In some example embodiments, the data collection initiation request may be sent from a group of first devices 110 (e.g., a UE) (303B). In some example embodiments, the data collection initiation request may be sent from a second device 120 (e.g., a network device) (303C).

[0130] In some example embodiments, a data collection initiation request may include a set of requirements for the data to be collected, indicating what kind of data to collect using a set of requirements (e.g., characteristics of the data to be collected). In some example embodiments, a data collection initiation request may also include at least one ID of at least one AI / ML model (referred to as a “Model ID (MID)”) for which data is to be collected. Where data collection is related to a specific Model ID, multiple IDs of (multiple) AI / ML models are included in the data collection initiation request. As an example, the characteristics of the data to be collected may request a set of RSRP samples to be collected periodically (e.g., every 20 milliseconds) for a set of beams indexed by a beam ID (e.g., an SSB index). The data characteristics may also indicate a threshold on the RSRP samples, e.g., all RSRP samples with RSRPs higher than -90 dBm. It should be understood that any other requirements for the data to be collected may be defined, and this is not limited to the scope of this disclosure.

[0131] In response to a data collection initiation request, a third device 130 assigns (312) a Temporary Model Identifier (TMID) to associate with data to be collected in an AI / ML-related data collection performed by a set of first devices 110. In some example embodiments, a TMID may be assigned when a model ID is not provided. In some example embodiments, if a model ID is provided, the model ID may be used as a TMID or may be mapped to an assigned TMID. The TMID may be used as a token to link the data to be collected. The third device 130 sends (310A, 310B, or 310C) the assigned TMID to the entity that sent the data collection initiation request. An external entity (e.g., an application service) may receive the TMID. In some cases, a set of first devices 110 may receive (306B) a TMID from the third device 130. In some cases, a second device 120 may receive (306C) a TMID from the third device 130.

[0132] The third device 130 sends a (314) data collection request to the second device 120. The data collection request includes characteristics of the data to be collected and an assigned TMID, which is used to associate the data to be collected in an AI / ML-related data collection performed by a set of first devices 110. In some example embodiments, the data collection request may also include a MID.

[0133] Upon receiving a data collection request (316) from the third device 130, the second device 120 sends a data collection token (DCT) (318) to the third device 130. The DCT can be assigned for AI / ML-related data collection and therefore can be used to associate the data collection request in step 314, since data collection is not an immediate action and may take some time before the network can complete data collection from one or more of the first devices 110. In some example embodiments, such as Figure 3A As shown, the DCT can be included in the data collection request acknowledgment (ACK) sent to the third device 130. In some example embodiments, the DCT can be sent to the third device 130 in association with the TMID.

[0134] After receiving (320) the DCT assigned by the second device 120, the third device 130 maps (322) the TMID assigned by the third device 130 for the requested AI / ML related data collection to the received DCT. The third device 130 can also track the data collection session.

[0135] The second device 120 can retrieve (324) the capability signaling of the first device 110. In some example embodiments, the first device 110 can send the capability signaling of the first device 110 to the second device 120. In some example embodiments, the second device 120 can retrieve the capability signaling of the first device 110 for a given set of TMIDs and MIDs. As described above, the capability signaling can include conditions for enabling AI / ML features and / or AI / ML-related data collection, for example, indicating whether the first device 110 supports enabled AI / ML features, and / or whether the first device 110 supports AI / ML-related data collection.

[0136] The second device 120 may expect the first device 110 to indicate multiple TMIDs and multiple MIDs associated with data collection, and the first device 110 to report its capability to support the data collection process used for model development. In some cases, some of the first devices 110 that support data collection may not support AI / ML-enabled features and may only act as data collection entities within the network. In some cases, once data has been collected at the first device 110, it can be assumed that the UE vendor-specific dataset is being delivered to the UE server. As mentioned, a first device 110 that does not support AI / ML-enabled features may still indicate certain model IDs or other indicators associated with conditions and additional conditions. AI / ML-related data collection features can be defined as features independent of other AI / ML features and may also involve more than one model ID.

[0137] The second device 120 controls the AI / ML related data collection at the first device 110 by configuring and instructing data collection process-related information in a dynamic manner, simultaneously considering multiple first devices 110. The second device 120 sends (326) configuration information for AI / ML related data collection to a group of first devices 110 (which may be selected based on retrieved capability signaling).

[0138] Upon receiving (328) configuration information for AI / ML related data collection, a group of first devices 110 initiates an AI / ML related data collection process based on the configuration information to perform data collection actions (330). The group of first devices 110 may send (332) a data collection request ACK to the second device 120 as a response to the data collection request. Upon receiving (334) the data collection request ACK, the second device 120 may determine that the group of first devices 110 has confirmed the data collection action.

[0139] A group of first devices 110 can be selected by an algorithm within the second device 120. For example, the second device 120 can group first devices 110 that report the same model ID and have a common capability regarding AI / ML-related data collection (e.g., the ability to collect beam level reference signal received power (RSRP) samples for up to 64 Synchronous SignAI / PBCH block (SSB) beam indices). The second device 120 can sort the configuration based on one or more sets of conditions reflecting possible changes in network configuration. For example, the second device 120 can switch from 16 SSB beams to 64 SSB beams during peak hours and switch back to 16 SSB beams due to energy savings during off-peak hours. Since this affects the AI / ML-related data collection process, it is important to track this on the first device 110 side when requesting AI / ML-related data collection. This is referred to as “network configuration conditions” or “condition set,” which can have some influence on AI / ML-related data collection on the first device 110 side. The first device 110 is expected to achieve this by mixing data samples without crossing different "network configuration conditions". Additional conditions are reported in the UE capability corresponding to (multiple) model IDs.

[0140] In some example embodiments, when network conditions remain stable under a first network configuration condition, the second device 120 may initiate a data collection process at the first device 110, wherein the first network configuration condition may correspond to additional conditions associated with the model(s)(s) model(s) initiated by the data collection process. In some example embodiments, when network configuration conditions change from the first network configuration condition to the second network configuration condition, the second device 120 may pause or stop the data collection process at the first device 110, wherein the second network configuration condition may not involve additional conditions associated with the model(s) of the ongoing data collection process. When network configuration conditions switch back to the first network configuration condition, the second device 120 may restart or initiate the data collection process at the first device 110.

[0141] In some example embodiments, configuration information associated with GC-PDCCH monitoring may be provided by the second device 110 prior to enabling or disabling the data collection process, whereby the GC-PDCCH is associated with AI / ML-related data collection performed by a set of first devices. For one or more first devices 110 supporting data collection for developing ML-enabled features, the first device 110 may receive configuration information enabling the monitoring of the GC-PDCCH associated with the data collection. In some examples, the configuration information associated with GC-PDCCH monitoring may be provided as part of the data collection configuration in the data collection request, or may be provided separately before or after the data collection request, whereby the GC-PDCCH is associated with AI / ML-related data collection.

[0142] Depending on the matching between network configuration conditions and conditions related to AI / ML related data collection, the second device 120 may send (336) indication information via GC-PDCCH associated with AI / ML related data collection, the indication information being used to enable AI / ML related data collection at a set of first devices 110.

[0143] The first device 110 can monitor the GC-PDCCH based on the received configuration information. After decoding (338) the GC-PDCCH associated with AI / ML related data collection, the first device 110 can decode one or more indicators from the GC-PDCCH. In the example, the indication information carried in the GC-PDCCH may include a first indicator indicating the enabling or disabling of the data collection process on the first device 110 side and a second indicator indicating the ID that can be associated with the ML model identified by the second device 120. If the first indicator indicates the enabling of the AI / ML related data collection process, the first device 110 can perform (340) actions on the GC-PDCCH and the indication information to initiate or resume AI / ML related data collection for the ML model identified by the second indicator.

[0144] In some cases, if the second device 120 determines to disable AI / ML related data collection, it further sends (342) indication information via the GC-PDCCH associated with the AI / ML related data collection, indicating the disabling of AI / ML related data collection at a set of first devices 110. After decoding (344) the GC-PDCCH associated with the AI / ML related data collection, the first device 110 may decode one or more indicators from the GC-PDCCH. If a first indicator indicates disabling the data collection process, the first device 110 may pause or stop AI / ML related data collection for the ML model identified by the second indicator.

[0145] In some example embodiments, the second device 120 may receive data collected by a set of first devices 110 and transmit the collected data to a third device 130 in association with DCT. In some example embodiments, if the second device 120 decides to terminate or suspend data collection at the first device 110 for an extended period, it may additionally provide the first device 110 with some uplink grant configuration to transmit the data buffer collected after the data collection process has been terminated.

[0146] The second device 120 can retrieve (346) a set of data buffers from the first device 110. In some example embodiments, the second device 120 uploads the data collected in the data buffers to the third device 130. The second device 120 can send the collected data in association with the DCT. After receiving (350) the collected data, the third device 130 can store (352) the collected data in association with the TMID mapped to the DCT. In some example embodiments, if the AI / ML related data collection process terminates, the second device 120 can release the DCT.

[0147] In some example embodiments, the third device 130 may also perform (354) some processing on the collected data, such as filtering, sample removal, outlier removal, etc. In some example embodiments, the third device 130 may deliver (356A, 356B, 356C) the processed data to the entity that sent the data collection initiation request. The entity requesting the data may then continue with model training or other processing related to LCM of the AI / ML model.

[0148] In some examples, data may be delivered to an external entity, such as an application service. In some examples, (multiple) first devices 110 may receive (358B) data. In some examples, a second device 120 may receive (358C) data. In some example embodiments, data may be delivered in association with a TMID, allowing an entity to associate the data with a previously sent data collection initiation request.

[0149] Figure 4 A flowchart of an example method 400 implemented at a first device according to some example embodiments of the present disclosure is shown. For the purposes of discussion, [the following will be discussed]. Figure 1 Method 400 is described by the angle of the first device 110 in the middle.

[0150] At box 410, the first device 110 receives configuration information from the second device 120 for monitoring the Group Common Physical Downlink Control Channel (GC-PDCCH), which is associated with artificial intelligence (AI) / machine learning (ML) related data collection.

[0151] At box 420, the first device 110 monitors the GC-PDCCH associated with AI / ML related data collection based on configuration information.

[0152] At box 430, based on the determination that GC-PDCCH is detected, the first device 110 determines from the detected GC-PDCCH an indication message for enabling or disabling AI / ML related data collection.

[0153] At frame 440, the first device 110 performs actions on the AI / ML related data collection process based on instruction information.

[0154] In some example embodiments, the configuration information for monitoring the GC-PDCCH indicates at least one monitoring instruction for the GC-PDCCH associated with AI / ML related data collection, the monitoring instruction indicating at least one of the following: a temporary identifier for a dedicated radio network for data collection, information related to the interpretation of downlink control information carried in the GC-PDCCH, configuration of at least one associated AI / ML function for AI / ML related data collection, or resource configuration for monitoring the GC-PDCCH.

[0155] In some example embodiments, the information related to the interpretation of the downlink control information carried in the GC-PDCCH indicates at least one of the following: the starting position of the information within the downlink control information carried in the GG-PDCCH, an identifier of a predefined downlink control information format, or a list of identifiers of AI / ML models associated with AI / ML-related data collection to be enabled or disabled.

[0156] In some example embodiments, the indication information includes a first indicator indicating the enabling or disabling of AI / ML related data collection and a second indicator indicating at least one identifier of at least one AI / ML model identified by a second device.

[0157] In some example implementations, the indication information is carried in the GC-PDCCH according to a predefined downlink control information format.

[0158] In some example embodiments, the second indicator includes a bitmap, where bits in the bitmap correspond to the identifier of the AI / ML model identified by the second device.

[0159] In some example embodiments, based on determining that a first indicator indicates the enabling of AI / ML related data collection, the first device 110 may initiate or resume the AI / ML related data collection process for at least one AI / ML model indicated by a second indicator. Based on determining that the first indicator indicates the disabling of AI / ML related data collection, the first device 110 may suspend or terminate the AI / ML related data collection process for at least one AI / ML model indicated by the second indicator.

[0160] In some example embodiments, the first device 110 may receive from the second device an assignment of at least one identifier of at least one AI / ML model identified by the second device, wherein the identifier of the AI / ML model is assigned to identify at least one of the following: a first condition reported by the first device in capability signaling for enabling AI / ML features, or a second condition not reported in capability signaling.

[0161] In some example embodiments, depending on whether the AI / ML related data collection process is to be paused or terminated, the first device 110 may receive resource authorization from the second device; and use the resource authorization to send data collected after the pause or termination of the AI / ML related data collection process to the second device.

[0162] In some example embodiments, the GC-PDCCH is configured to be monitored by a group of first devices, including a first device.

[0163] In some example embodiments, the first device 110 may send a data collection initiation request to a third device that manages AI / ML related data collection, the data collection initiation request including a request for data to be collected; and receive from the third device a temporary model identifier for associating the data to be collected in the AI / ML related data collection performed by the first device.

[0164] In some example embodiments, the data collection initiation request also includes at least one identifier for at least one AI / ML model for which data is to be collected.

[0165] In some example embodiments, the first device 110 may send capability signaling of the first device to the second device; receive configuration information for AI / ML related data collection from the second device; and initiate an AI / ML related data collection process based on the configuration information for AI / ML related data collection.

[0166] In some example embodiments, the capability signaling indicates at least one of the following: an indication of whether the AI / ML enabled feature is supported by the first device, an indication of whether AI / ML related data collection is supported by the first device, or at least one identifier of at least one AI / ML model associated with AI / ML related data collection supported by the first device.

[0167] In some example embodiments, the first device includes a terminal device, and the second device includes a network device.

[0168] Figure 5 A flowchart of an example method 500 implemented at a second device according to some example embodiments of the present disclosure is shown. For the purposes of discussion, [the following will be discussed]. Figure 1 Method 500 is described by the angle of the second device 120 in the middle.

[0169] At box 510, the second device 120 sends configuration information associated with monitoring of the Group Common Physical Downlink Control Channel (GC-PDCCH) to a group of first devices 110. The GC-PDCCH is associated with artificial intelligence (AI) / machine learning (ML) related data collection performed by the group of first devices.

[0170] At frame 520, the second device 120 determines whether AI / ML related data collection should be enabled or disabled.

[0171] At box 530, the second device 120 sends an instruction message via GC-PDCCH to a group of first devices to enable or disable AI / ML related data collection, based on a determination of whether AI / ML related data collection should be enabled or disabled.

[0172] In some example embodiments, the configuration information for monitoring GC-PDCCH indicates at least one monitoring instruction for GC-PDCCH associated with AI / ML related data collection, the monitoring instruction indicating at least one of the following: a temporary identifier for a dedicated radio network for data collection, information related to the interpretation of downlink control information (DCI) carried in GC-PDCCH, configuration of at least one associated AI / ML function for AI / ML related data collection, or resource configuration for monitoring GC-PDCCH.

[0173] In some example embodiments, the indication information includes a first indicator indicating the enabling or disabling of AI / ML related data collection and a second indicator indicating at least one identifier of at least one AI / ML model identified by a second device.

[0174] In some example embodiments, the second device is made to determine whether AI / ML related data collection should be enabled or disabled based on a match between at least one condition of at least one AI / ML model associated with AI / ML related data collection and network configuration conditions for sending signals that the data collected by a set of first devices is transmitted.

[0175] In some example embodiments, based on determining that a first network configuration condition matches at least one condition of at least one AI / ML model, the second device 120 may send a first indication message via GC-PDCCH to a group of first devices to enable an AI / ML related data collection process for at least one AI / ML model; and based on determining that the first network configuration condition changes to a second network configuration condition that does not match at least one condition of at least one AI / ML model, the second device 120 may send a second indication message via GC-PDCCH to a group of first devices to disable the AI / ML related data collection process for at least one AI / ML model.

[0176] In some example embodiments, the second device 120 may send to the first device an assignment of at least one identifier of at least one AI / ML model identified by the second device, wherein the identifier of the AI / ML model is assigned to identify at least one of the following: a first condition reported by the first device in capability signaling for enabling AI / ML features, or a second condition not reported in capability signaling.

[0177] In some example embodiments, depending on whether the AI / ML related data collection process is to be paused or terminated, the second device 120 may send a resource grant to the first device; and use the resource grant to receive data collected from the first device after the AI / ML related data collection process has been paused or terminated.

[0178] In some example embodiments, the second device 120 may receive a data collection request from a third device that manages AI / ML related data collection. The data collection request includes characteristics of the data to be collected and a temporary model identifier for associating the data to be collected in AI / ML related data collection performed by a set of first devices; and sends a data collection token assigned to the third device for AI / ML related data collection.

[0179] In some example embodiments, the second device 120 may receive data collected by a group of first devices; and send the collected data to a third device in association with a data collection token.

[0180] In some example embodiments, the second device 120 may receive capability signaling from at least one set of first devices; select a first set of devices for AI / ML related data collection based at least in part on the capability signaling of at least one set of first devices; and send configuration information for AI / ML related data collection to the set of first devices.

[0181] In some example embodiments, the second device 120 may send a data collection initiation request to a third device that manages AI / ML related data collection, the data collection initiation request including a request for data to be collected; and receive from the third device a temporary model identifier for associating data to be collected in AI / ML related data collection performed by a set of first devices.

[0182] In some example embodiments, the data collection initiation request also includes at least one identifier for at least one AI / ML model for which data is to be collected.

[0183] Figure 6 A flowchart of an example method 600 implemented at a third device according to some example embodiments of the present disclosure is shown. For the purposes of discussion, [the following will be discussed]. Figure 1 Method 600 is described by the angle of the third device 130 in the middle.

[0184] At box 610, in response to a data collection initiation request, the third device 130 assigns a temporary model identifier to associate with the data to be collected in an AI / ML-related data collection performed by a set of first devices.

[0185] At box 620, the third device 130 sends a data collection request to the second device 120. The data collection request includes characteristics of the data to be collected and temporary model identifiers for associating the data to be collected in an AI / ML-related data collection performed by a set of first devices.

[0186] At frame 630, the third device 130 receives a data collection token assigned to the second device for AI / ML related data collection.

[0187] At box 640, the third device 130 maps the data collection token to a temporary model identifier.

[0188] In some example embodiments, the third device 130 may receive a data collection request from at least one of the following: a first set of devices, a second device, or an external entity.

[0189] In some example embodiments, the data collection initiation request also includes at least one identifier for at least one AI / ML model for which data is to be collected. The third device 130 may map at least one identifier of the at least one AI / ML model to a temporary model identifier.

[0190] In some example embodiments, the third device 130 may receive data collected by a group of first devices from the second device, the collected data being received in association with a data collection token; and store the collected data in association with a temporary model identifier mapped to the data collection token.

[0191] In some example embodiments, the third device 130 may process the collected data and deliver the processed data to the entity that sent the data collection initiation request.

[0192] In some example embodiments, the first means capable of performing any step of method 400 (e.g., Figure 1 The first device 110 may include a component for performing a corresponding operation of method 400. This component may be implemented in any suitable form. For example, the component may be implemented in a circuit or software module. The first device may be implemented as or included in... Figure 1 In the first device 110.

[0193] In some example embodiments, the first device includes: components for receiving configuration information from the second device for monitoring a Group Common Physical Downlink Control Channel (GC-PDCCH), the GC-PDCCH being associated with artificial intelligence (AI) / machine learning (ML) related data collection; components for monitoring the GC-PDCCH associated with the AI / ML related data collection based on the configuration information; components for determining, based on the determination that the GC-PDCCH is detected, indication information for enabling or disabling the AI / ML related data collection from the detected GC-PDCCH; and components for performing actions on the AI / ML related data collection process based on the indication information.

[0194] In some example embodiments, the configuration information for monitoring GC-PDCCH indicates at least one monitoring instruction for GC-PDCCH associated with AI / ML related data collection, the monitoring instruction indicating at least one of the following: a temporary identifier for a dedicated radio network for data collection, information related to the interpretation of downlink control information carried in GC-PDCCH, configuration of at least one associated AI / ML function for AI / ML related data collection, or resource configuration for monitoring GC-PDCCH.

[0195] In some example embodiments, the information related to the interpretation of the downlink control information carried in the GC-PDCCH indicates at least one of the following: the starting position of the information within the downlink control information carried in the GG-PDCCH, an identifier of a predefined downlink control information format, or a list of identifiers of AI / ML models associated with AI / ML-related data collection to be enabled or disabled.

[0196] In some example embodiments, the indication information includes a first indicator indicating the enabling or disabling of AI / ML related data collection and a second indicator indicating at least one identifier of at least one AI / ML model identified by a second device.

[0197] In some example implementations, the indication information is carried in the GC-PDCCH according to a predefined downlink control information format.

[0198] In some example embodiments, the second indicator includes a bitmap, where bits in the bitmap correspond to the identifier of the AI / ML model identified by the second device.

[0199] In some example embodiments, the components for performing actions on the AI / ML related data collection process based on indication information include: components for initiating or resuming the AI / ML related data collection process for at least one AI / ML model indicated by a second indicator based on determining that the first indicator indicates the activation of AI / ML related data collection; and components for suspending or terminating the AI / ML related data collection process for at least one AI / ML model indicated by the second indicator based on determining that the first indicator indicates the deactivation of AI / ML related data collection.

[0200] In some example embodiments, the first device further includes: a component for receiving from the second device the assignment of at least one identifier of at least one AI / ML model identified by the second device, wherein the identifier of the AI / ML model is assigned to identify at least one of the following: a first condition reported by the first device in capability signaling for enabling AI / ML features, or a second condition not reported in capability signaling.

[0201] In some example embodiments, the first device further includes: a component for receiving resource authorization from the second device based on determining that the AI / ML related data collection process is to be paused or terminated; and a component for sending data collected after the pause or termination of the AI / ML related data collection process to the second device using the resource authorization.

[0202] In some example embodiments, the GC-PDCCH is configured to be monitored by a group of first devices, including a first device.

[0203] In some example embodiments, the first device further includes: a component for sending a data collection initiation request to a third device for managing AI / ML related data collection, the data collection initiation request including a request for data to be collected; and a component for receiving from the third device a temporary model identifier for associating data to be collected in an AI / ML related data collection performed by a set of first devices.

[0204] In some example embodiments, the data collection initiation request also includes at least one identifier for at least one AI / ML model for which data is to be collected.

[0205] In some example embodiments, the first device further includes: a component for sending capability signaling of the first device to the second device; a component for receiving configuration information for AI / ML related data collection from the second device; and a component for initiating an AI / ML related data collection process based on the configuration information for AI / ML related data collection.

[0206] In some example embodiments, the capability signaling indicates at least one of the following: an indication of whether the AI / ML enabled feature is supported by the first device, an indication of whether AI / ML related data collection is supported by the first device, or at least one identifier of at least one AI / ML model associated with AI / ML related data collection supported by the first device.

[0207] In some example embodiments, the first device includes a terminal device, and the second device includes a network device.

[0208] In some example embodiments, the first device further includes components for performing additional operations in some example embodiments of method 400 or first device 110. In some example embodiments, the components include at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause execution of the first device.

[0209] In some example embodiments, a second device capable of performing any step of method 500 (e.g., Figure 1 The second device 120 may include a component for performing the corresponding operation of method 500. This component may be implemented in any suitable form. For example, the component may be implemented in a circuit or software module. The second device may be implemented as or included in... Figure 1 The second device 120 in the middle.

[0210] In some example embodiments, the second device includes: components for transmitting configuration information associated with monitoring of a group common physical downlink control channel (GC-PDCCH) to a group of first devices, the GC-PDCCH being associated with artificial intelligence (AI) / machine learning (ML) related data collection performed by the group of first devices; components for determining whether AI / ML related data collection should be enabled or disabled; and components for transmitting indication information for enabling or disabling AI / ML related data collection to the group of first devices via the GC-PDCCH based on the determination that AI / ML related data collection should be enabled or disabled.

[0211] In some example embodiments, the configuration information for monitoring GC-PDCCH indicates at least one GC-PDCCH associated with AI / ML related data collection, and the monitoring instructions indicate at least one of the following: a temporary identifier for a dedicated radio network for data collection, information related to the interpretation of downlink control information (DCI) carried in the GC-PDCCH, the configuration of at least one AI / ML function associated with AI / ML related data collection, or the resource configuration for monitoring GC-PDCCH.

[0212] In some example embodiments, the indication information includes a first indicator indicating the enabling or disabling of AI / ML related data collection and a second indicator indicating at least one identifier of at least one AI / ML model identified by a second device.

[0213] In some example embodiments, the component for determining whether AI / ML related data collection should be enabled or disabled includes: a component for determining whether AI / ML related data collection should be enabled or disabled based on a match between at least one condition of at least one AI / ML model associated with AI / ML related data collection and network configuration conditions for transmitting signals collected by a set of first devices.

[0214] In some example embodiments, the components for determining whether AI / ML related data collection should be enabled or disabled based on matching include: components for sending, via GC-PDCCH, first indication information for enabling the AI / ML related data collection process for at least one AI / ML model to a group of first devices based on determining that a first network configuration condition matches at least one condition of at least one AI / ML model; and components for sending, via GC-PDCCH, second indication information for disabling the AI / ML related data collection process for at least one AI / ML model to a group of first devices based on determining that the first network configuration condition changes to a second network configuration condition that does not match at least one condition of at least one AI / ML model.

[0215] In some example embodiments, the second device further includes: a component for sending to the first device an assignment of at least one identifier of at least one AI / ML model identified by the second device, wherein the identifier of the AI / ML model is assigned to identify at least one of the following: a first condition reported by the first device in capability signaling for enabling AI / ML features, or a second condition not reported in capability signaling.

[0216] In some example embodiments, the second device further includes: a component for sending a resource authorization to the first device based on determining that the AI / ML related data collection process should be suspended or terminated; and a component for receiving data collected from the first device after the suspension or termination of the AI / ML related data collection process using the resource authorization.

[0217] In some example embodiments, the second device further includes: a component for receiving a data collection request from a third device that manages AI / ML related data collection, the data collection request including characteristics of the data to be collected and a temporary model identifier for associating the data to be collected in an AI / ML related data collection to be performed by a set of first devices; and a component for sending a data collection token assigned to the third device for AI / ML related data collection.

[0218] In some example embodiments, the second device further includes: a component for receiving data collected by a group of first devices; and a component for sending the collected data to a third device in association with a data collection token.

[0219] In some example embodiments, the second device further includes: components for receiving capability signaling of at least one set of first devices from at least one set of first devices; components for selecting a first set of first devices for AI / ML related data collection based at least in part on the capability signaling of at least one set of first devices; and components for sending configuration information for AI / ML related data collection to the set of first devices.

[0220] In some example embodiments, the second device further includes: a component for sending a data collection initiation request to a third device for managing AI / ML related data collection, the data collection initiation request including a request for data to be collected; and a component for receiving from the third device a temporary model identifier for associating data to be collected in an AI / ML related data collection performed by a set of first devices.

[0221] In some example embodiments, the data collection initiation request also includes at least one identifier for at least one AI / ML model for which data is to be collected.

[0222] In some example embodiments, the second device further includes components for performing additional operations in some example embodiments of method 500 or second device 120. In some example embodiments, the components include at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause execution of the second device.

[0223] In some example embodiments, a third device capable of performing any step of method 600 (e.g., Figure 1The third device 130 may include a component for performing the corresponding operation of method 600. This component may be implemented in any suitable form. For example, the component may be implemented in a circuit or software module. The third device may be implemented as or included in... Figure 1 The third device 130 in the middle.

[0224] In some example embodiments, the third device includes: means for allocating a temporary model identifier for associating data to be collected in an AI / ML-related data collection performed by a set of first devices in response to a data collection initiation request; means for sending a data collection request to a second device, the data collection request including characteristics of the data to be collected and a temporary model identifier for associating the data to be collected in an AI / ML-related data collection performed by a set of first devices; means for receiving a data collection token assigned to the AI / ML-related data collection from the second device; and means for mapping the data collection token to the temporary model identifier.

[0225] In some example embodiments, the third device further includes a component for receiving a data collection initiation request from at least one of the following: a first set of devices, a second device, or an external entity.

[0226] In some example embodiments, the data collection initiation request also includes at least one identifier of at least one AI / ML model for which data is to be collected, and the third device further includes a component for mapping at least one identifier of at least one AI / ML model to a temporary model identifier.

[0227] In some example embodiments, the third device further includes: a component for receiving data collected by a group of first devices from the second device, the collected data being received in association with a data collection token; and a component for storing the collected data in association with a temporary model identifier mapped to the data collection token.

[0228] In some example embodiments, the third device further includes: a component for processing the collected data; and a component for delivering the processed data to the entity that sent the data collection initiation request.

[0229] In some example embodiments, the third device further includes components for performing additional operations in some example embodiments of method 600 or third device 130. In some example embodiments, the components include at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause execution of the third device.

[0230] Figure 7 This is a simplified block diagram of a device 700 suitable for implementing exemplary embodiments of the present disclosure. Device 700 can be provided to implement a communication device, such as... Figure 1 The first device 110, the second device 120, or the third device 130 are shown. As shown, the device 700 includes one or more processors 710, one or more memories 720 coupled to the processors 710, and one or more communication modules 740 coupled to the processors 710.

[0231] Communication module 740 is used for bidirectional communication. Communication module 740 has one or more communication interfaces to support communication with one or more other modules or devices. The communication interface can represent any interface necessary for communication with other network elements. In some example embodiments, communication module 740 may include at least one antenna.

[0232] As a non-limiting example, processor 710 can be any type suitable for a local technology network and can include one or more of the following: general-purpose computer, special-purpose computer, microprocessor, digital signal processor (DSP), and processor based on a multi-core processor architecture. Device 700 can have multiple processors, such as application-specific integrated circuit chips that are time-dependent on a clock that synchronizes with the main processor.

[0233] Memory 720 may include one or more non-volatile memories and one or more volatile memories. Examples of non-volatile memories include, but are not limited to, read-only memory (ROM) 724, electrically programmable read-only memory (EPROM), flash memory, hard disk, compact disc (CD), digital video disc (DVD), optical disc, laser disc, and other magnetic and / or optical storage. Examples of volatile memories include, but are not limited to, random access memory (RAM) 722 and other volatile memories that will not persist during power-off periods.

[0234] Computer program 730 includes computer-executable instructions that are executed by an associated processor 710. The instructions of program 730 may include instructions for performing operations / actions of some example embodiments of this disclosure. Program 730 may be stored in memory (e.g., ROM 724). Processor 710 can perform any suitable actions and processes by loading program 730 into RAM 722.

[0235] The exemplary embodiments of this disclosure can be implemented by program 730, enabling device 700 to perform as described in the reference. Figures 2 to 6 Any process discussed in this disclosure. Exemplary embodiments of this disclosure may also be implemented by hardware or a combination of software and hardware.

[0236] In some example embodiments, program 730 may be tangibly contained in a computer-readable medium, which may be included in device 700 (such as in memory 720) or other storage devices accessible by device 700. Device 700 may load program 730 from the computer-readable medium into RAM 722 for execution. In some example embodiments, the computer-readable medium may include any type of non-transitory storage medium, such as ROM, EPROM, flash memory, hard disk, CD, DVD, etc. As used herein, the term "non-transitory" refers to a limitation of the medium itself (i.e., tangible, not tactile), rather than a limitation of the persistence of data storage (e.g., RAM versus ROM).

[0237] Figure 8 An example of a computer-readable medium 800 is shown, which may be in the form of a CD, DVD, or other optical storage disc. The computer-readable medium 800 has a program 730 stored thereon.

[0238] In some example embodiments, the following provides possible ways to implement solutions according to some example embodiments of this disclosure in the 3GPP communication specifications.

[0239] TS 38.212 (DCI format can be defined) The DCI format X_Y is used to notify changes in data collection for one or more UEs.

[0240] The following information is transmitted via DCI format X_Y with CRC scrambled by DC-RNTI: Block number 1, block number 2, ..., block number N The starting position of the block is provided by a parameter from the higher layer of the UE that configures the block. DC-PositionDCI-XY Sure.

[0241] If the UE is configured with higher-level parameters dc-RNTI and dci-FormatX-Y Then, a higher layer configures a block for the UE, defining the following fields for that block: Data collection initialization / stop indication -1 bit Auxiliary information indicator -0 bit, if no higher-level parameters are configured. model-ID-for- DataCollection; otherwise, it depends on the parameters of the higher level. model-ID-for- DataCollection provides different model-ID The number of bits determines a 1, 2, 3, 4, or L-bit bitmap, where each bit corresponds to one of the model IDs configured by higher-level parameters, and the MSB to LSB of the bitmap is arranged in order. model-IDThe ascending order corresponds to the model IDs configured from the first to the last.

[0242] The size of X_Y in DCI format is determined by higher-level parameters. sizeDCI-XY …instruct.

[0243] TS 38.213 (Can capture PDCCH monitoring) For UEs configured with AI / ML data collection, the following detection of DCI format X_Y in PDCCH reception can be provided. For passing dc-RNTI DC-RNTI of DCI format X_Y pass dci-FormatX-Y The monitoring of PDCCH is used to detect the number of search space sets of DCI format X_Y on the active DL BWP of PCell or SpCell according to the public search space as described below. For passing sizeDCI-X- DCI format X_Y payload size pass DC-PositionDCI-XY Position of the data collection initialization / stop indicator bit in DCI format X_Y A value of '0' for the data collection initialization / stop indicator bit indicates the start of the data collection process. A value of '1' for the data collection initialization / stop indicator bit indicates that the data collection process has stopped. Bitmap, when configuring higher-layer parameters of the UE model-ID-for-DataCollection At that time, among them The bitmap position immediately follows the data collection initialization / stop indicator bit. The bitmap size is equal to the size of the bitmap. model-ID-for-DataCollection Configuration in model-ID The number of bits in the bitmap, where each bit corresponds to the model ID. The '0' value of a bit used in a bitmap indicates that... model-ID-for-DataCollection Provided model-ID It is effective for the data collection process. - The '1' value of the bit used in the bitmap indicates the value determined by... model-ID-for-DataCollection Provided model-ID The data collection process is ineffective. ……… TS 38.331 (You can describe the RRC parameters used.) DC-PositionDCI-X -Y The starting position of the data collection initialization / stop indication in the DCI format XY (see TS 38.213…).

[0244] dci-FormatX-Y If configured, the UE monitors the DCI format X_Y according to TS 38.213 (clause…).

[0245] model-ID-for-DataCollection This field contains the model ID associated with data collection enabled / disabled by the DCI format XY. model-ID The model IDs are derived from a list of model IDs identified by the network via the UE capability identifier, and are configured for use in the data collection process at the UE. Configured in model-ID-for-DataCollection Inside.

[0246] Generally, the various embodiments of this disclosure can be implemented in hardware or dedicated circuitry, software, logic, or any combination thereof. Some aspects can be implemented in hardware, and others can be implemented in firmware or software that can be executed by a controller, microprocessor, or other computing device. Although various aspects of the embodiments of this disclosure are shown and described as block diagrams, flowcharts, or using some other graphical representation, it should be understood that the blocks, apparatuses, systems, techniques, or methods described herein can be implemented in hardware, software, firmware, dedicated circuitry or logic, general-purpose hardware, or controllers or other computing devices, or some combination thereof, as non-limiting examples.

[0247] Some exemplary embodiments of this disclosure also provide at least one computer program product tangibly stored on a computer-readable medium, such as a non-transitory computer-readable medium. The computer program product includes computer-executable instructions that execute in a device on a target physical or virtual processor, such as those included in a program module, to perform any step of the methods described above. Typically, a program module includes routines, programs, libraries, objects, classes, components, data structures, etc., that perform a particular task or implement a particular abstract data type. In various embodiments, the functionality of a program module can be combined or split among program modules as needed. The machine-executable instructions for a program module can execute within a local or distributed device. In a distributed device, the program module can reside on both local and remote storage media.

[0248] Program code used to perform the methods of this disclosure may be written in any combination of one or more programming languages. The program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that, when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a stand-alone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0249] In the context of this disclosure, computer program code or related data may be carried by any suitable carrier wave to enable a device, apparatus, or processor to perform the various processes and operations described above. Examples of carrier waves include signals, computer-readable media, etc.

[0250] Computer-readable media can be computer-readable signal media or computer-readable storage media. Computer-readable media can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any suitable combination thereof. More specific examples of computer-readable storage media will include electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable optical disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0251] Furthermore, although operations are described in a specific order, this should not be construed as requiring that such operations be performed in the specific order shown or sequentially, or requiring that all shown operations be performed to achieve the desired result. In some cases, multitasking and parallel processing may be advantageous. Similarly, although several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of this disclosure, but rather as a description of features that may be specific to particular embodiments. Unless explicitly stated otherwise, certain features described in the context of a single embodiment may also be implemented in combination in a single embodiment. Conversely, unless explicitly stated otherwise, various features described in the context of a single embodiment may also be implemented individually or in any suitable sub-combination in multiple embodiments.

[0252] Although this disclosure has been described in language specific to structural features and / or methodological actions, it should be understood that the disclosure as defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are disclosed as exemplary forms for implementing the claims.

Claims

1. A first device, comprising: At least one processor; as well as At least one memory stores instructions that, when executed by the at least one processor, cause the first device to perform at least the following: Configuration information for monitoring the Group Common Physical Downlink Control Channel (GC-PDCCH) is received from the second device, the GC-PDCCH being associated with artificial intelligence (AI) / machine learning (ML) related data collection; Based on the configuration information, monitor the GC-PDCCH associated with the AI / ML related data collection; Based on the determination that the GC-PDCCH is detected, an indication message for enabling or disabling the collection of AI / ML related data is determined from the detected GC-PDCCH; as well as Actions are performed on the AI / ML related data collection process based on the indicated information.

2. The apparatus of claim 1, wherein the configuration information for monitoring the GC-PDCCH indicates at least one monitoring instruction for the GC-PDCCH associated with the AI / ML related data collection, the monitoring instruction indicating at least one of the following: Temporary identifier for dedicated radio networks for data collection. Information related to the interpretation of the downlink control information carried in the GC-PDCCH. Configuration of at least one associated AI / ML function for the AI / ML related data collection, or Resource configuration for monitoring the GC-PDCCH.

3. The apparatus of claim 2, wherein the information relating to the interpretation of the downlink control information carried in the GC-PDCCH indicates at least one of the following: The indication information indicates the starting position within the downlink control information carried in the GG-PDCCH. Identifiers for a predefined downlink control information format, or A list of identifiers for AI / ML models associated with the AI / ML-related data collection to be enabled or disabled.

4. The apparatus according to any one of claims 1 to 3, wherein the indication information includes a first indicator indicating the activation or deactivation of the AI / ML related data collection and a second indicator indicating at least one identifier of at least one AI / ML model identified by the second apparatus.

5. The apparatus of claim 4, wherein the indication information is carried in the GG-PDCCH according to a predefined downlink control information format.

6. The apparatus of claim 4 or 5, wherein the second indicator comprises a bitmap, wherein bits in the bitmap correspond to an identifier of an AI / ML model identified by the second apparatus.

7. The apparatus according to any one of claims 4 to 6, wherein the first apparatus is caused to perform: Based on the determination that the first indicator indicates the enabling of AI / ML related data collection, initiate or resume the AI / ML related data collection process for the at least one AI / ML model indicated by the second indicator; and Based on the determination that the first indicator indicates the deactivation of the AI / ML related data collection, the AI / ML related data collection process for the at least one AI / ML model indicated by the second indicator is suspended or terminated.

8. The apparatus according to any one of claims 1 to 7, wherein the first apparatus is further caused to perform: Receive assignment of at least one identifier of at least one AI / ML model identified by the second device. The identifier for the AI / ML model is assigned to identify at least one of the following: The first condition reported in the capability signaling of the first device for enabling AI / ML features, or A second condition not reported in the capability signaling.

9. The apparatus according to any one of claims 1 to 8, wherein the first apparatus is further caused to perform: Based on the determination that the AI / ML related data collection process needs to be paused or terminated, resource authorization is received from the second device; and The resource authorization is used to send the data collected after the AI / ML related data collection process was paused or terminated to the second device.

10. The apparatus according to any one of claims 1 to 9, wherein the GC PDCCH is configured to be monitored by a group of first devices including the first device.

11. The apparatus according to any one of claims 1 to 10, wherein the first apparatus is further caused to perform: Sending a data collection initiation request to a third device that manages the collection of AI / ML related data, the data collection initiation request including a requirement for the data to be collected; and The third device receives a temporary model identifier for associating with data to be collected in the AI / ML related data collection performed by the first device.

12. The apparatus of claim 11, wherein the data collection initiation request further includes at least one identifier for at least one AI / ML model for which the data is to be collected.

13. The apparatus according to any one of claims 1 to 12, wherein the first apparatus is further caused to perform: Send the capability signaling of the first device to the second device; Receive configuration information for the AI / ML related data collection from the second device; and Based on the configuration information for collecting the AI / ML related data, the AI / ML related data collection process is initiated.

14. The apparatus of claim 13, wherein the capability signaling indicates at least one of the following: An indication of whether the AI / ML feature is supported by the first device. Instructions regarding whether the collection of AI / ML-related data is supported by the first device, or At least one identifier of at least one AI / ML model associated with the AI / ML related data collection supported by the first device.

15. The apparatus according to any one of claims 1 to 14, wherein the first apparatus comprises a terminal device and the second apparatus comprises a network device.

16. A second device, comprising: At least one processor; as well as At least one memory stores instructions that, when executed by the at least one processor, cause the second device to perform at least the following: Configuration information associated with monitoring of a Group Common Physical Downlink Control Channel (GC-PDCCH) is sent to a group of first devices, the GC-PDCCH being associated with artificial intelligence (AI) / machine learning (ML) related data collection performed by the group of first devices; Determine whether the AI / ML related data collection should be enabled or disabled; and Based on the determination of whether the AI / ML related data collection should be enabled or disabled, the GC-PDCCH sends an instruction message for enabling or disabling the AI / ML related data collection to the group of first devices.

17. The second apparatus of claim 16, wherein the configuration information for monitoring the GC-PDCCH indicates at least one monitoring instruction for the GC-PDCCH associated with the AI / ML related data collection, the monitoring instruction indicating at least one of the following: Temporary identifier for dedicated radio networks for data collection. Information related to the interpretation of the downlink control information (DCI) carried in the GC-PDCCH. Configuration of at least one associated AI / ML function for the AI / ML related data collection, or Resource configuration for monitoring the GC-PDCCH.

18. The second device according to claim 16 or 17, wherein the indication information includes a first indicator indicating the activation or deactivation of the AI / ML related data collection and a second indicator indicating at least one identifier of at least one AI / ML model identified by the second device.

19. The second device according to any one of claims 16 to 18, wherein the second device is caused to perform: Based on the matching between at least one condition of at least one AI / ML model associated with the AI / ML related data collection and network configuration conditions for transmitting signals that the data collected by the group of first devices is transmitted, it is determined whether the AI / ML related data collection should be enabled or disabled.

20. The second apparatus of claim 19, wherein the second apparatus is caused to perform: Based on the determination that a first network configuration condition matches at least one condition of the at least one AI / ML model, a first indication message for enabling the AI / ML related data collection process for the at least one AI / ML model is sent via the GC-PDCCH to the group of first devices; and Based on the determination that the first network configuration condition has been changed to a second network configuration condition that does not match the at least one condition of the at least one AI / ML model, a second instruction message for disabling the AI / ML related data collection process for the at least one AI / ML model is sent to the group of first devices via the GC-PDCCH.

21. The second device according to any one of claims 16 to 20, wherein the second device is further caused to perform: Send an assignment to the first device of at least one identifier for at least one AI / ML model identified by the second device. The identifier for the AI / ML model is assigned to identify at least one of the following: The first condition reported in the capability signaling of the first device for enabling AI / ML features, or A second condition not reported in the capability signaling.

22. The second device according to any one of claims 16 to 21, wherein the second device is further caused to perform: Based on the determination that the AI / ML related data collection process needs to be paused or terminated, a resource authorization is sent to the first device; and The resource authorization is used to receive data collected from the first device after the AI / ML related data collection process has been paused or terminated.

23. The second device according to any one of claims 16 to 22, wherein the second device is further caused to perform: A data collection request is received from a third device managing AI / ML-related data collection, the data collection request including characteristics of the data to be collected and a temporary model identifier for associating the data to be collected in the AI / ML-related data collection performed by the set of first devices; and Send a data collection token, assigned for the collection of AI / ML related data, to the third device.

24. The second apparatus of claim 23, wherein the second apparatus is further caused to perform: Receive data collected by the set of first devices; and The collected data is sent to the third device in association with the data collection token.

25. The second device according to any one of claims 16 to 24, wherein the second device is further caused to perform: Receive capability signaling from at least the group of first devices; The set of first devices for collecting the AI / ML related data is selected, at least in part, based on the capability signaling of at least the set of first devices; and Send configuration information for the collection of AI / ML related data to the first set of devices.

26. The second device according to any one of claims 16 to 25, wherein the second device is further caused to perform: Sending a data collection initiation request to a third device that manages the collection of AI / ML related data, the data collection initiation request including a requirement for the data to be collected; and The third device receives a temporary model identifier for associating with data to be collected in the AI / ML related data collection performed by the set of first devices.

27. The second apparatus of claim 26, wherein the data collection initiation request further includes at least one identifier for at least one AI / ML model for which the data is to be collected.

28. A third device, comprising: At least one processor; as well as At least one memory stores instructions that, when executed by the at least one processor, cause the third device to perform at least the following: In response to a data collection initiation request, a temporary model identifier is assigned to associate the data to be collected in an AI / ML-related data collection performed by a set of first devices; A data collection request is sent to a second device, the data collection request including characteristics of the data to be collected and a temporary model identifier for associating the data to be collected in the AI / ML related data collection performed by the set of first devices; Receive a data collection token assigned to the AI / ML related data collection from the second device; as well as Map the data collection token to the temporary model identifier.

29. The third device according to claim 28, wherein the third device is further caused to perform: The data collection initiation request is received from at least one of the following: the first set of devices, the second device, or an external entity.

30. The third apparatus of claim 28 or 29, wherein the data collection initiation request further includes at least one identifier for at least one AI / ML model for which the data is to be collected, and the third apparatus is further caused to perform: Map at least one identifier of at least one AI / ML model to the temporary model identifier.

31. The third device according to any one of claims 28 to 30, wherein the third device is further caused to perform: Receive, from the second device, the data collected by the group of first devices, wherein the collected data is received in association with the data collection token; and The collected data is stored in association with the temporary model identifier mapped to the data collection token.

32. The third device according to claim 31, wherein the third device is further caused to perform: Process the collected data; and The processed data is delivered to the entity that sent the data collection initiation request.

33. A method comprising: At the first device, configuration information for monitoring the Group Common Physical Downlink Control Channel (GC-PDCCH) is received from the second device, which is associated with artificial intelligence (AI) / machine learning (ML) related data collection; Based on the configuration information, monitor the GC-PDCCH associated with the AI / ML related data collection; Based on the determination that the GC-PDCCH is detected, an indication message for enabling or disabling the collection of AI / ML related data is determined from the detected GC-PDCCH; as well as Actions are performed on the AI / ML related data collection process based on the indicated information.

34. A method comprising: At the second device, configuration information associated with the monitoring of the Group Common Physical Downlink Control Channel (GC-PDCCH) is sent to a group of first devices, the GC-PDCCH being associated with artificial intelligence (AI) / machine learning (ML) related data collection performed by the group of first devices; Determine whether the AI / ML related data collection should be enabled or disabled; and Based on the determination of whether the AI / ML related data collection should be enabled or disabled, the GC-PDCCH sends an instruction message for enabling or disabling the AI / ML related data collection to the group of first devices.

35. A method comprising: In response to a data collection initiation request, a temporary model identifier is allocated at a third device to associate the data to be collected in an AI / ML-related data collection performed by a set of first devices; Send a data collection request to the second device, the data collection request including the characteristics of the data to be collected and the temporary model identifier for associating the data to be collected in the AI / ML related data collection performed by the set of first devices; Receive a data collection token assigned to the AI / ML related data collection from the second device; as well as Map the data collection token to the temporary model identifier.

36. A first device, comprising: A component for receiving configuration information from a second device for monitoring the Group Common Physical Downlink Control Channel (GC-PDCCH), which is associated with AI / ML related data collection for the GC-PDCCH; A component used to monitor the GC-PDCCH associated with the AI / ML related data collection based on the configuration information; A component for determining, based on the detection of the GC-PDCCH, an indication of enabling or disabling the collection of AI / ML related data from the detected GC-PDCCH; as well as A component for performing actions on the AI / ML related data collection process based on the indicated information.

37. A second device, comprising: A component for transmitting configuration information associated with monitoring of a group common physical downlink control channel (GC-PDCCH) to a group of first devices, the GC-PDCCH being associated with artificial intelligence (AI) / machine learning (ML) related data collection performed by the group of first devices; Components used to determine whether the collection of AI / ML related data should be enabled or disabled; as well as A component for sending an instruction message for enabling or disabling AI / ML related data collection to the group of first devices via the GC-PDCCH based on the determination that AI / ML related data collection should be enabled or disabled.

38. A third device, comprising: A component for allocating a temporary model identifier for associating with data to be collected in an AI / ML-related data collection performed by a set of first devices in response to a data collection initiation request; A component for sending a data collection request to a second device, the data collection request including characteristics of the data to be collected and a temporary model identifier for associating the data to be collected in the AI / ML related data collection performed by the set of first devices; A component for receiving, from the second device, a data collection token assigned for the collection of AI / ML related data; as well as A component used to map the data collection token to the temporary model identifier.

39. A computer-readable medium having instructions stored thereon for causing a device to perform at least the method of claim 33, the method of claim 34, or the method of claim 35.

Images