Capturing additional conditions

Abstract

The present disclosure relates to capturing additional conditions. The method includes at a first apparatus, receiving, from a second apparatus, model feature information comprising a plurality of first training configurations identified by respective first configuration identifiers, wherein the plurality of first training configurations is 5 associated with communication condition; determining, for each first training configuration, a plurality of second training configurations identified by respective second configuration identifiers; obtaining, for each second training configuration, a plurality of datasets associated with communication condition based on the plurality of first training configurations, wherein one dataset is obtained based on one first training configuration 0 from the plurality of first training configuration; and transmitting, to the second apparatus, data collection information comprising the plurality of second training configurations identified by respective second configuration identifiers.

Description

CAPTURING ADDITIONAL CONDITIONSFIELDS

[0001] This application claims priority to, and the benefit of, GB Application No. 2402173.5, filed February 16, 2024, the contents of which are hereby incorporated by reference in their entirety.FIELDS

[0002] Various example embodiments of the present disclosure generally relate to the field of telecommunication and in particular, to methods, devices, apparatuses and computer readable storage medium for capturing additional conditions.BACKGROUND

[0003] With developments in the integration of Artificial Intelligence (Al) and Machine Learning (ML) within the 5G and emerging 6G New Radio (NR) Air Interface, a new frontier in network adaptability and efficiency is being explored. The 3GPP Release-18 study emphasizes the importance of model generalization across various network scenarios, addressing the need for AI / ML models to maintain robust performance under diverse conditions. This includes the strategic incorporation of additional conditions to refine model training, ensuring models are well-suited to both network-side and user equipment-side requirements. Therefore, it is worth delving into the aspects of capturing additional conditions, as understanding and leveraging these factors are crucial for the evolution of AI / ML applications in 5G and beyond, with the potential to significantly impact network performance and user experience.SUMMARY

[0004] In a first aspect of the present disclosure, there is provided a first apparatus. The first apparatus comprises at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the first apparatus to: receive, from a second apparatus, model feature information comprising a plurality of first training configurations identified by respective first configuration identifiers, wherein the plurality of first training configurations is associated with communication condition; determine, for each first training configuration, a plurality of second training iconfigurations identified by respective second configuration identifiers; obtain, for each second training configuration, a plurality of datasets associated with communication condition based on the plurality of first training configurations, wherein one dataset is obtained based on one first training configuration from the plurality of first training configuration; and transmit, to the second apparatus, data collection information comprising the plurality of second training configurations identified by respective second configuration identifiers.

[0005] In a second aspect of the present disclosure, there is provided a second apparatus. The second apparatus comprises at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the second apparatus to: transmit, to a first apparatus, model feature information comprising a plurality of first training configurations identified by respective first configuration identifiers, wherein the plurality of first training configurations is associated with communication condition; for each first training configuration, receive, from the first apparatus, data collection information comprising a plurality of second training configurations identified by respective second configuration identifiers; and obtain, for each second training configuration, a plurality of datasets associated with communication condition based on the plurality of second training configurations, wherein one dataset is obtained based on one second training configuration from the plurality of second training configuration.

[0006] In a third aspect of the present disclosure, there is provided a method. The method comprises: receiving, from a second apparatus, model feature information comprising a plurality of first training configurations identified by respective first configuration identifiers, wherein the plurality of first training configurations is associated with communication condition; determining, for each first training configuration, a plurality of second training configurations identified by respective second configuration identifiers; obtaining, for each second training configuration, a plurality of datasets associated with communication condition based on the plurality of first training configurations, wherein one dataset is obtained based on one first training configuration from the plurality of first training configuration; and transmitting, to the second apparatus, data collection information comprising the plurality of second training configurations identified by respective second configuration identifiers.

[0007] In a fourth aspect of the present disclosure, there is provided a method. The method comprises: transmitting, to a first apparatus, model feature information comprising a plurality of first training configurations identified by respective firstconfiguration identifiers, wherein the plurality of first training configurations is associated with communication condition; for each first training configuration, receiving, from the first apparatus, data collection information comprising a plurality of second training configurations identified by respective second configuration identifiers; and obtaining, for each second training configuration, a plurality of datasets associated with communication condition based on the plurality of second training configurations, wherein one dataset is obtained based on one second training configuration from the plurality of second training configuration.

[0008] In a fifth aspect of the present disclosure, there is provided a first apparatus. The first apparatus comprises means for receiving, from a second apparatus, model feature information comprising a plurality of first training configurations identified by respective first configuration identifiers, wherein the plurality of first training configurations is associated with communication condition; means for determining, for each first training configuration, a plurality of second training configurations identified by respective second configuration identifiers; means for obtaining, for each second training configuration, a plurality of datasets associated with communication condition based on the plurality of first training configurations, wherein one dataset is obtained based on one first training configuration from the plurality of first training configuration; and means for transmitting, to the second apparatus, data collection information comprising the plurality of second training configurations identified by respective second configuration identifiers.

[0009] In a sixth aspect of the present disclosure, there is provided a second apparatus. The second apparatus comprises means for transmitting, to a first apparatus, model feature information comprising a plurality of first training configurations identified by respective first configuration identifiers, wherein the plurality of first training configurations is associated with communication condition; means for each first training configuration, receiving, from the first apparatus, data collection information comprising a plurality of second training configurations identified by respective second configuration identifiers; and means for obtaining, for each second training configuration, a plurality of datasets associated with communication condition based on the plurality of second training configurations, wherein one dataset is obtained based on one second training configuration from the plurality of second training configuration.

[0010] In a seventh aspect of the present disclosure, there is provided a computer readable medium. The computer readable medium comprises instructions stored thereon for causing an apparatus to perform at least the method according to the third aspect.

[0011] In an eighth aspect of the present disclosure, there is provided a computer readable medium. The computer readable medium comprises instructions stored thereon for causing an apparatus to perform at least the method according to the fourth aspect.

[0012] It is to be understood that the Summary section is not intended to identify key or essential features of embodiments of the present disclosure, nor is it intended to be used to limit the scope of the present disclosure. Other features of the present disclosure will become easily comprehensible through the following description.BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Some example embodiments will now be described with reference to the accompanying drawings, where:

[0014] FIG. 1 illustrates an example communication environment in which example embodiments of the present disclosure can be implemented;

[0015] FIG. 2A and FIG. 2B illustrate a signaling flow of additional conditions grouping in accordance with some example embodiments of the present disclosure;

[0016] FIG. 3 illustrates a signaling flow of user equipment side and grouping of additional conditions in accordance with some example embodiments of the present disclosure;

[0017] FIG. 4 illustrates a signaling flow of model training and identification using the information in network and user equipment sided additional conditions grouping in accordance with some example embodiments of the present disclosure;

[0018] FIG. 5 illustrates a flowchart of a method implemented at a first device according to some example embodiments of the present disclosure;

[0019] FIG. 6 illustrates a flowchart of a method implemented at a second device according to some example embodiments of the present disclosure;

[0020] FIG. 7 illustrates a simplified block diagram of a device that is suitable for implementing example embodiments of the present disclosure; and

[0021] Throughout the drawings, the same or similar reference numerals represent the same or similar element.DETAILED DESCRIPTION

[0022] Principle of the present disclosure will now be described with reference to some example embodiments. It is to be understood that these embodiments are described only for the purpose of illustration and help those skilled in the art to understand and implement the present disclosure, without suggesting any limitation as to the scope of the disclosure.Embodiments described herein can be implemented in various manners other than the ones described below.

[0023] In the following description and claims, unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skills in the art to which this disclosure belongs.

[0024] References in the present disclosure to “one embodiment,” “an embodiment,” “an example embodiment,” and the like indicate that the embodiment described may include a particular feature, structure, or characteristic, but it is not necessary that every embodiment includes the particular feature, structure, or characteristic. Moreover, such phrases are not necessarily referring to the same embodiment. Further, when a particular feature, structure, or characteristic is described in connection with an embodiment, it is submitted that it is within the knowledge of one skilled in the art to affect such feature, structure, or characteristic in connection with other embodiments whether or not explicitly described.

[0025] It shall be understood that although the terms “first,” “second,”... , etc. in front of noun(s) and the like may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another and they do not limit the order of the noun(s). For example, a first element could be termed a second element, and similarly, a second element could be termed a first element, without departing from the scope of example embodiments. As used herein, the term “and / or” includes any and all combinations of one or more of the listed terms.

[0026] As used herein, “at least one of the following: ” and “at least one of ” and similar wording, where the list of two or more elements are joined by “and” or “or”, mean at least any one of the elements, or at least any two or more of the elements, or at least all the elements.

[0027] As used herein, unless stated explicitly, performing a step “in response to A” does not indicate that the step is performed immediately after “A” occurs and one or more intervening steps may be included.

[0028] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments. As used herein, the singular forms “a”, “an” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms “comprises”, “comprising”, “has”, “having”, “includes” and / or “including”, when used herein, specify the presence of stated features, elements, and / or components etc., butdo not preclude the presence or addition of one or more other features, elements, components and / or combinations thereof.

[0029] As used in this application, the term “circuitry” may refer to one or more or all of the following:(a) hardware-only circuit implementations (such as implementations in only analog and / or digital circuitry) and(b) combinations of hardware circuits and software, such as (as applicable):(i) a combination of analog and / or digital hardware circuit(s) with software / firmware and(ii) any portions of hardware processor(s) with software (including digital signal processor(s)), software, and memory(ies) that work together to cause an apparatus, such as a mobile phone or server, to perform various functions) and(c) hardware circuit(s) and or processor(s), such as a microprocessor(s) or a portion of a microprocessor(s), that requires software (e.g., firmware) for operation, but the software may not be present when it is not needed for operation.

[0030] This definition of circuitry applies to all uses of this term in this application, including in any claims. As a further example, as used in this application, the term circuitry also covers an implementation of merely a hardware circuit or processor (or multiple processors) or portion of a hardware circuit or processor and its (or their) accompanying software and / or firmware. The term circuitry also covers, for example and if applicable to the particular claim element, a baseband integrated circuit or processor integrated circuit for a mobile device or a similar integrated circuit in server, a cellular network device, or other computing or network device.

[0031] As used herein, the term “communication network” refers to a network following any suitable communication standards, such as New Radio (NR), Long Term Evolution (LTE), LTE-Advanced (LTE-A), Wideband Code Division Multiple Access (WCDMA), High-Speed Packet Access (HSPA), Narrow Band Internet of Things (NB-IoT), wireless local-area network (WLAN), WiFi and so on. Furthermore, the communications between a terminal device and a network device in the communication network may be performed according to any suitable generation communication protocols, including, but not limited to, the first generation (1G), the second generation (2G), 2.5G, 2.75G, the third generation (3G), the fourth generation (4G), 4.5G, the fifth generation (5G), the sixth generation (6G)communication protocols, and / or any other protocols either currently known or to be developed in the future. Embodiments of the present disclosure may be applied in various communication systems. Given the rapid development in communications, there will of course also be future type communication technologies and systems with which the present disclosure may be embodied. It should not be seen as limiting the scope of the present disclosure to only the aforementioned system.

[0032] As used herein, the term “network device” refers to a node in a communication network via which a terminal device accesses the network and receives services therefrom. The network device may refer to a base station (BS) or an access point (AP), for example, a node B (NodeB or NB), an evolved NodeB (eNodeB or eNB), an NR NB (also referred to as a gNB), a Remote Radio Unit (RRU), a radio header (RH), a remote radio head (RRH), an Operations and Maintenance entity (0AM) entity, a Positioning Reference Unit (PRU), Location Management Function (LMF), a relay, an Integrated Access and Backhaul (IAB) node, a low power node such as a femto, a pico, a non-terrestrial network (NTN) or non-ground network device such as a satellite network device, a low earth orbit (LEO) satellite and a geosynchronous earth orbit (GEO) satellite, an aircraft network device, and so forth, depending on the applied terminology and technology. In some example embodiments, radio access network (RAN) split architecture comprises a Centralized Unit (CU) and a Distributed Unit (DU) at an IAB donor node. An IAB node comprises a Mobile Terminal (IAB-MT) part that behaves like a UE toward the parent node, and a DU part of an IAB node behaves like a base station toward the next-hop IAB node.

[0033] The term “terminal device” refers to any end device that may be capable of wireless communication. By way of example rather than limitation, a terminal device may also be referred to as a communication device, user equipment (UE), a Subscriber Station (SS), a Portable Subscriber Station, a Mobile Station (MS), or an Access Terminal (AT). The terminal device may include, but not limited to, a mobile phone, a cellular phone, a smart phone, voice over IP (VoIP) phones, wireless local loop phones, a tablet, a wearable terminal device, a personal digital assistant (PDA), portable computers, desktop computer, image capture terminal devices such as digital cameras, gaming terminal devices, music storage and playback appliances, vehicle-mounted wireless terminal devices, wireless endpoints, mobile stations, laptop-embedded equipment (LEE), laptop-mounted equipment (LME), USB dongles, smart devices, wireless customer-premises equipment (CPE), an Internet of Things (loT) device, a watch or other wearable, a head-mounteddisplay (HMD), a vehicle, a drone, a medical device and applications (e.g., remote surgery), an industrial device and applications (e.g., a robot and / or other wireless devices operating in an industrial and / or an automated processing chain contexts), a consumer electronics device, a device operating on commercial and / or industrial wireless networks, and the like. The terminal device may also correspond to a Mobile Termination (MT) part of an IAB node (e.g., a relay node). In the following description, the terms “terminal device”, “communication device”, “terminal”, “user equipment” and “UE” may be used interchangeably.

[0034] As used herein, the term “resource,” “transmission resource,” “resource block,” “physical resource block” (PRB), “uplink resource,” or “downlink resource” may refer to any resource for performing a communication, for example, a communication between a terminal device and a network device, such as a resource in time domain, a resource in frequency domain, a resource in space domain, a resource in code domain, or any other combination of the time, frequency, space and / or code domain resource enabling a communication, and the like. In the following, unless explicitly stated, a resource in both frequency domain and time domain will be used as an example of a transmission resource for describing some example embodiments of the present disclosure. It is noted that example embodiments of the present disclosure are equally applicable to other resources in other domains.

[0035] As used herein, the term "model feature information" may refer to the specific characteristics or attributes in a computational model, particularly in the fields of machine learning and statistical analysis. These features represent the key factors that the model analyzes to make predictions, perform classifications, or execute other data-driven tasks. It is noted that example embodiments of the present disclosure are equally applicable to model feature information in other domains.

[0036] As used herein, the term "communication condition" may refer to a current state or quality of a communication link or network, encompassing various factors such as positioning, signal strength, noise levels, bandwidth availability, and interference. It is noted that example embodiments of the present disclosure are equally applicable to communication conditions in other domains.

[0037] As used herein, the term "training configuration" may refer to the systematic setup and parameters defined for gathering and preparing data specifically for developing and refining machine learning algorithms within the communication sector. The training configuration may also include machine learning (ML)-specific training parameters, e.g.ML model hyper-parameters, ML input data pre-processing or ML output data postprocessing configurations, and so on. The ML-specific training parameters are critical for predicting outcomes accurately. For example, in optimizing network performance through predictive analytics, a data collection configuration might involve the strategic capture of network load patterns, error rates, and service quality indicators at different times and locations to train models that can predict and mitigate potential network issues. It is noted that example embodiments of the present disclosure are equally applicable to data collection configurations in other domains.

[0038] As used herein, the term "configuration identifier" may refer to a unique label or code that distinguishes a specific set of parameters or settings used during the data collection, preprocessing, and model training processes. This identifier enables the systematic tracking, replication, and comparison of different machine learning experiments or training sessions. For example, when training models to predict network congestion, a configuration identifier might be used to denote a particular combination of features extracted from network traffic data. It is noted that example embodiments of the present disclosure are equally applicable to configuration identifiers in other domains.

[0039] As used herein, the term " buffer" may refer to a temporary storage area, typically in memory, used to hold data while it is being transferred from one place to another, ensuring smooth and efficient data processing. A buffer may be used to manage the flow of data packets across a network, accommodating variations in data transmission rates and preventing packet loss during high-traffic conditions. It is noted that example embodiments of the present disclosure are equally applicable to buffers in other domains.

[0040] As used herein, the term "characteristic differences" refers to the distinct variations or discrepancies between data configurations, or datasets in terms of their attributes, properties, or behaviors. For example, in the field of machine learning, identifying characteristic differences between data points is essential for training algorithms to distinguish between categories or predict outcomes accurately. It is noted that example embodiments of the present disclosure are equally applicable to identifying characteristic differences in other domains.

[0041] As used herein, the term "core network device" refers to the essential hardware components within a telecommunications network that provide key functionalities for data routing, management, and connectivity across the network. These devices facilitate the central operations of the network, supporting data transmission between various parts of the network and to external networks. It is noted that example embodiments of the presentdisclosure are equally applicable to core network devices in other domains.

[0042] As used herein, the term "radio access network device" refers to the components and equipment within a telecommunications system that connect mobile devices to the core network and facilitate wireless communication. These devices are integral to establishing and maintaining the radio link between the user's mobile device and the network, handling tasks such as signal transmission, reception, and modulation. It is noted that example embodiments of the present disclosure are equally applicable to radio access network devices in other domains.

[0043] As used herein, a machine learning (ML) entity may contain an ML model and ML model related metadata. The ML entity may be managed as a single composite entity. In some example embodiments, the ML entity may be implemented as a MLApp.

[0044] To facilitate understanding of the terminologies, RANI agreements on the list of terminologies used for AI / ML are provided below.

[0045] AI / ML Model: A data driven algorithm that applies AI / ML techniques to generate a set of outputs based on a set of inputs.

[0046] AI / ML model delivery: A generic term referring to delivery of an AI / ML model from one entity to another entity in any manner. Note: An entity could mean a network node / function (e.g., gNB, LMF, etc.), UE, proprietary server, etc.

[0047] AI / ML model Inference: A process of using a trained AI / ML model to produce a set of outputs based on a set of inputs.

[0048] AI / ML model testing: A subprocess of training, to evaluate the performance of a final AI / ML model using a dataset different from one used for model training and validation. Differently from AI / ML model validation, testing does not assume subsequent tuning of the model.

[0049] AI / ML model training: A process to train an AI / ML Model [by learning the input / output relationship] in a data driven manner and obtain the trained AI / ML Model for inference.

[0050] AI / ML model transfer: Delivery of an AI / ML model over the air interface in a manner that is not transparent to 3GPP signalling, either parameters of a model structure known at the receiving end or a new model with parameters. Delivery may contain a full model or a partial model.

[0051] AI / ML model validation: A subprocess of training, to evaluate the quality of an AI / ML model using a dataset different from one used for model training, that helps selecting model parameters that generalize beyond the dataset used for model training.

[0052] Data collection: A process of collecting data by the network nodes, management entity, or UE for the purpose of AI / ML model training, data analytics and inference.

[0053] Federated learning / federated training: A machine learning technique that trains an AI / ML model across multiple decentralized edge nodes (e.g., UEs, gNBs) each performing local model training using local data samples. The technique requires multiple interactions of the model, but no exchange of local data samples.

[0054] Functionality identification: A process / method of identifying an AI / ML functionality for the common understanding between the NW and the UE. Note: Information regarding the AI / ML functionality may be shared during functionality identification. Where AI / ML functionality resides depends on the specific use cases and sub use cases.

[0055] Model activation: enable an AI / ML model for a specific function.

[0056] Model deactivation: disable an AI / ML model for a specific function.

[0057] Model download: Model transfer from the network to UE.

[0058] Model identification: A process / method of identifying an AI / ML model for the common understanding between the NW and the UE. Note: The process / method of model identification may or may not be applicable. Note: Information regarding the AI / ML model may be shared during model identification.

[0059] Model monitoring: A procedure that monitors the inference performance of the AI / ML model.

[0060] Model parameter update: Process of updating the model parameters of a model.

[0061] Model selection: The process of selecting an AI / ML model for activation among multiple models for the same AI / ML enabled feature. Note: Model selection may or may not be carried out simultaneously with model activation.

[0062] Model switching: Deactivating a currently active AI / ML model and activating a different AI / ML model for a specific function.

[0063] Model update: Process of updating the model parameters and / or model structure of a model.

[0064] Model upload: Model transfer from UE to the network.

[0065] Network-side (AI / ML) model: An AI / ML Model whose inference is performed entirely at the network.

[0066] Offline field data: The data collected from field and used for offline training of the AI / ML model.

[0067] Offline training: An AI / ML training process where the model is trained basedon collected dataset, and where the trained model is later used or delivered for inference. Note: This definition only serves as a guidance. There may be cases that may not exactly conform to this definition but could still be categorized as offline training by commonly accepted conventions.

[0068] Online field data: The data collected from field and used for online training of the AI / ML model.

[0069] Online training: An AI / ML training process where the model being used for inference) is (typically continuously) trained in (near) real-time with the arrival of new training samples. Note: the notion of (near) real-time vs. non real-time is context- dependent and is relative to the inference time-scale. Note: This definition only serves as a guidance. There may be cases that may not exactly conform to this definition but could still be categorized as online training by commonly accepted conventions. Note: Fine- tuning / re-training may be done via online or offline training. (This note could be removed when we define the term fine-tuning.)

[0070] Reinforcement Learning (RL): A process of training an AI / ML model from input (a.k.a. state) and a feedback signal (a.k.a. reward) resulting from the model’s output (a.k.a. action) in an environment the model is interacting with.

[0071] Semi-supervised learning: A process of training a model with a mix of labelled data and unlabelled data.

[0072] Supervised learning: A process of training a model from input and its corresponding labels.

[0073] Two-sided (AI / ML) model: A paired AI / ML Model(s) over which joint inference is performed, where joint inference comprises AI / ML Inference whose inference is performed jointly across the UE and the network, i.e, the first part of inference is firstly performed by UE and then the remaining part is performed by gNB, or vice versa.

[0074] UE-side (AI / ML) model: An AI / ML Model whose inference is performed entirely at the UE.

[0075] Unsupervised learning: A process of training a model without labelled data.

[0076] Proprietary-format models: ML models of vendor-Zdevice-specific proprietary format, from 3GPP perspective. They are not mutually recognizable across vendors and hide model design information from other vendors when shared. Note: An example is a device-specific binary executable format.

[0077] Open-format models: ML models of specified format that are mutually recognizable across vendors and allow interoperability, from 3GPP perspective. They aremutually recognizable between vendors and do not hide model design information from other vendors when shared.

[0078] In 3GPP Release-18 study on Artificial Intelligence (AI)ZMachine Learning (ML) for NR Air Interface, several use-cases were considered, and evaluation studies were conducted to check the generalization performance of the used ML models, In particular for channel state information (CSI) feedback enhancements with ML models, for Model generalization performance, various scenarios / configurations and combinations thereof are used during training and validation phase to ensure model can perform across variety of scenarios / configurations. Some of these parameters are specific to the simulation scenario (e.g., UE distribution), some of these parameters are known based on system information (e.g., carrier frequency) and some of them are specific to product implementation (e.g., antenna panel structure and orientation) and maybe even specific to RF aspects e.g., power used for a given transmission bandwidth.

[0079] The set of scenarios are considered focusing on one or more of the following aspects:-Various deployment scenarios (e.g., urban micro (UMi), universal mobile access (UMa), indoor hotspot (InH));-Various outdoor / indoor UE distributions for UMa / UMi (e.g., 10:0, 8:2, 5:5, 2:8, 0: 10);-Various carrier frequencies (e.g., 2GHz, 3.5GHz);-Other aspects of scenarios are not precluded, e.g., various antenna spacing, various antenna virtualization (TxRU mapping), various ISDs, various UE speeds, etc;-Approach of dataset mixing across scenarios / configurations.

[0080] Over various configurations (e.g., which may potentially lead to different dimensions of model input / output), the set of configurations are considered focusing on one or more of the following aspects:-Various bandwidths (e.g., 10MHz, 20MHz) and / or frequency granularities, (e.g., size of sub-band);-Various sizes of CSI feedback payloads;-Various antenna port layouts, e.g., (Nl(the number of antenna ports) / N2 / P) and / or antenna port numbers (e.g., 32 ports, 16 ports);-Various UE speeds (e.g., lOkm / h, 30km / h, 60km / h, 120km / h, etc.) for CSI prediction sub use case;-Other aspects of configurations are not precluded, e.g., various numerologies, various rank numbers / layers, etc.

[0081] In further discussions, some of the above aspects were referred as “Additional Conditions”. For an AI / ML-enabled feature / FG, additional conditions may refer to any aspects that are assumed for the training of the model but are not a part of UE capability for the AI / ML-enabled feature / FG. It does not imply that additional conditions are necessarily specified. Additional conditions can be divided into two categories: NW-side additional conditions and UE-side additional conditions.

[0082] These additional conditions may be used in the discussions of model identification and model-ID-LCM, where:• For AI / ML model identification and model-ID-based life cycle management (LCM) of UE-side models and / or UE-part of two-sided models, model-ID-based LCM operates based on identified models, where a model may be associated with specific configurations / conditions associated with UE capability of an AI / ML- enabled Feature / FG and additional conditions (e.g., scenarios, sites, and datasets) as determined / identified between UE-side and NW-side.• From RANI perspective, an AI / ML model identified by a model ID may be logical, and how it maps to physical AI / ML model(s) may be up to implementation. When distinction is necessary for discussion purposes, companies may use the term a logical AI / ML model to refer to a model that is identified and assigned a model ID, and physical AI / ML model(s) to refer to an actual implementation of such a model.

[0083] RANI has made further clarified model identification in technical report (TR) 38.843 as below.0084] The term “network (NW) side additional condition” used herein may refer to a condition at network side which may potentially lead to different dimensions of model input / output. According to the above definition of model identification, the process of model identification pinpoints what are known as "additional conditions", which may include, for example, training dataset category, site-related information, timestamps, implicit identification information (such as, labels for specific gNB / UE implementation details), statistical information (e.g., delay spread, angular spread, LOS / NLOS data and so on), and other factors. These may also include model hyperparameters which control how a neural network learns from data. These hyperparameters affect the neural network architecture, the optimization, and the regularization (controlling overfitting or underfitting) of the underlying model. For Convolutional Neural Networks (CNN’s), hyperparameters include the size of kernels, number of kernels, length of strides, and pooling size, which directly affect the performance and training speed of CNNs and the set of selected hyperparameters impact the model training as more of these parameters increases as the complexity of the network increases and performance of the model varies widely based on such choice. However, how to accurately describe these additional conditions is a challenging task.

[0085] FIG. 1 illustrates an example communication environment 100 in which example embodiments of the present disclosure can be implemented. In the communication environment 100, two communication apparatus, including a first apparatus 110, and a second apparatus 120 can communicate with each other.

[0086] In the example of FIG. 1, the first apparatus 110 may be a terminal device, such as UE, and the second apparatus 120 may be a network device, such as a core network device or a radio access network device. The serving area of the second apparatus 120 may be called a cell 102. The communication link 130 between the first apparatus 110 and the second apparatus 120 is over the air.

[0087] It is to be understood that the number of apparatuses and their connections shown in FIG. 1 are only for the purpose of illustration without suggesting any limitation. The communication environment 100 may include any suitable number of apparatuses configured to implementing example embodiments of the present disclosure. Althoughnot shown, it would be appreciated that one or more additional apparatuses may be located in the cell 102, and one or more additional cells may be deployed in the communication environment 100. In some example embodiments, both the first apparatus 110 and the second apparatus 120 may be UEs. Alternatively, both the first apparatus 110 and the second apparatus 120 may be radio network devices. In some other example embodiments, both the first apparatus 110 and the second apparatus 120 may be network devices.

[0088] In the following, for the purpose of illustration, some example embodiments are described with the first apparatus 110 operating as a terminal device and the second apparatus 120 operating as a network device. However, in some example embodiments, operations described in connection with a terminal device may be implemented at a network device or other device, and operations described in connection with a network device may be implemented at a terminal device or other device.

[0089] In some example embodiments, if the first apparatus 110 is a terminal device and the second apparatus 120 is a network device, a link from the second apparatus 120 to the first apparatus 110 is referred to as a downlink (DL), and a link from the first apparatus 110 to the second apparatus 120 is referred to as an uplink (UL). In DL, the second apparatus 120 is a transmitting (TX) device (or a transmitter) and the first apparatus 110 is a receiving (RX) device (or a receiver). In UL, the first apparatus 110 is a TX device (or a transmitter) and the second apparatus 120 is a RX device (or a receiver).

[0090] Communications in the communication environment 100 may be implemented according to any proper communication protocol(s), comprising, but not limited to, cellular communication protocols of the first generation (1G), the second generation (2G), the third generation (3G), the fourth generation (4G), the fifth generation (5G), the sixth generation (6G), and the like, wireless local network communication protocols such as Institute for Electrical and Electronics Engineers (IEEE) 802.11 and the like, and / or any other protocols currently known or to be developed in the future. Moreover, the communication may utilize any proper wireless communication technology, comprising 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 (OFDM), Discrete Fourier Transform spread OFDM (DFT-s-OFDM) and / or any other technologies currently known or to be developed in the future.

[0091] According to some example embodiments of the present disclosure, there isprovided a solution for the complex challenge of managing network-side and user equipment-side (UE-side) additional conditions in the context of Artificial Intelligence (Al) and Machine Learning (ML) model training and deployment. Traditional methods face difficulties due to the lack of transparency in encoding these conditions, complicating the task for data processors like UEs or Over-The-Top (OTT) servers to discern compatibility among different conditions. The present disclosure introduces a standardized protocol for the identification, encoding, and communication of these conditions. Such a system ensures that AI / ML models and data samples are precisely matched with the respective network and UE environments, aiming to significantly enhance the accuracy and efficiency of AI / ML applications within the network, thereby improving user experiences by adapting services to the distinct characteristics of each UE.

[0092] Example embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.

[0093] Reference is made to FIG. 2A and FIG. 2B, which illustrates a signaling flow including two portions 200-1 and 200-2 of additional conditions grouping in accordance with some example embodiments of the present disclosure. For the purpose of discussion, the signaling flow will be discussed with reference to FIG. 1, for example, by using the first apparatus 110 and the second apparatus 120. The signaling flow described with reference to FIG. 2A and FIG. 2B may be applied in difference scenarios, for example, beam management, channel state information (CSI) estimation, or positioning.

[0094] As shown in FIG. 2A, the second apparatus 120 transmits (2005) model feature information to the first apparatus 110. In other words, the first apparatus 110 receives (2005) model feature information from the second apparatus 120. The model feature information includes a plurality of first training configurations identified by respective first configuration identifiers, and the plurality of first training configurations is associated with communication condition. For example, the first training configuration may include one or more of: a scenario, a reference signal configuration, a data measurement configuration, or a data log configuration. The first configuration identifier may be any proper kind of information that can identify the first training configuration. In some example embodiments, the model feature information may include any characteristics or attributes in a computational model, particularly in the fields of machine learning and statistical analysis. It is noted that the model feature information may include any other proper model related information.

[0095] The first apparatus 110 determines (2010) a plurality of second trainingconfigurations for each first training configuration. The second training configuration is identified by respective second configuration identifiers. For example, the second training configuration may include one or more of: a scenario, a reference signal configuration, a data measurement configuration, or a data log configuration.

[0096] The first apparatus 110 obtains (2015) a plurality of datasets associated with communication condition for each second training configuration. The communication condition is based on the plurality of first training configurations, and one dataset is obtained based on one first training configuration from the plurality of first training configuration. In some embodiment, the communication condition may include reference signal transmissions. Alternatively, or in addition, the communication condition may include a beam management configuration. In some other embodiments, the communication condition may include positioning configuration.

[0097] The first apparatus 110 transmits (2020) data collection information to the second apparatus 120. In other words, the second apparatus 120 receives (2020) data collection information from the first apparatus 110. The data collection information includes the plurality of second training configurations identified by respective second configuration identifiers.

[0098] The second apparatus 120 obtains (2025) a plurality of datasets associated with communication condition for each second training configuration. The communication condition is based on the plurality of second training configurations, and one dataset is obtained based on one second training configuration from the plurality of second training configuration.

[0099] In some example embodiments, the first apparatus 110 stores (2030) the plurality of datasets in a buffer. The buffer may store data according to training configuration. In this case, each dataset is stored together with a corresponding first configuration identifier and a corresponding second configuration identifier. In some example embodiments, the second apparatus 120 stores (2032) the plurality of datasets in a buffer, and each dataset is stored together with a corresponding first configuration identifier and a corresponding second configuration identifier. It is noted the order of storing 2030 and 2032 in FIG. 2A is only an example.

[0100] In some example embodiments, the second apparatus 120 transmits (2035) a first indication to the first apparatus 110. In other words, the first apparatus 110 receives (2035) the first indication from the second apparatus 120. The first indication may indicate a set of first configuration identifiers, and characteristic differences among a set of first trainingconfigurations identified by respective the set of first configuration identifiers are smaller than a threshold. For example, if the characteristic differences are smaller than the threshold, it means that the set of first training configurations are compatible.

[0101] In some example embodiments, the first apparatus 110 determines (2040) a set of second configuration identifiers. In this case, a set of second training configurations identified by respective the set of second configuration identifiers may be compatible with each other. In some other example embodiments, the first apparatus 110 determines (2045) a set of datasets from the plurality of datasets, and the datasets are based on the set of first configuration identifiers and the set of second configuration identifiers. In some example embodiments, the first apparatus 110 stores (2050) the set of datasets together with the set of first configuration identifiers and the set of second configuration identifiers in the buffer. In some other example embodiments, the first apparatus 110 determines (2055) a data identity of the set of datasets. In some example embodiments, the first apparatus 110 trains a first model based on the set of datasets.

[0102] In some other example embodiments, the first apparatus 110 transmits (2060) a second indication to the second apparatus 120. In other words, the second apparatus 120 receives (2060) the second indication from the first apparatus 110. The second indication may indicate the set of second configuration identifiers.

[0103] In some example embodiments, the second apparatus 120 determines (2061) a set of datasets from the plurality of datasets based on the set of first configuration identifiers and the set of second configuration identifiers. In some other example embodiments, the second apparatus 120 stores (2062) the set of datasets together with the set of first configuration identifiers and the set of second configuration identifiers in the buffer which stores data according to the training configuration. In some example embodiments, the second apparatus 120 determines (2063) a dataset identity of the set of datasets. In some example embodiments, the second apparatus 120 trains (2064) a first model based on the set of datasets.

[0104] As shown in FIG. 2B, in some example embodiments, the first apparatus 110 receives (2065) a model transfer request from the second apparatus 120. In other words, the second apparatus 120 transmits (2065) the model transfer request to the first apparatus 110. The model transfer request may include a group of second configuration identifiers.

[0105] In some example embodiments, the first apparatus 110 determines (2070) a subgroup of second configuration identifiers from the group of second configuration identifiers. In this case, the second training configurations identified by respective thesubgroup of second configuration identifiers are compatible.

[0106] In some example embodiments, the first apparatus 110 transmits (2075) a mode transfer response to the second apparatus 120. In other words, the second apparatus 120 receives (2075) a mode transfer response from the first apparatus 110. The mode transfer response may indicate the subgroup of second configuration identifiers.

[0107] In some example embodiments, the first apparatus 110 receives (2080) a second model from the second apparatus 120. In other words, the second apparatus 120 transmits (2080) the second model to the first apparatus 110. The second model may be trained at the second apparatus 120 based on training configurations with the subgroup of second configuration identifiers.

[0108] In some example embodiments, the first apparatus 110 receives (2085) a model transfer request from the OTT server 210. In other words, the OTT server 210 transmits (2085) the model transfer request to the first apparatus 110. The model transfer request may include a group of second configuration identifiers.

[0109] In some example embodiments, the first apparatus 110 determines (2090) a subgroup of second configuration identifiers from the group of second configuration identifiers. In this case, the second training configurations identified by respective the subgroup of second configuration identifiers are compatible.

[0110] In some example embodiments, the first apparatus 110 transmits (2095) a mode transfer response to the OTT server 210. In other words, the OTT server 210 receives (2095) a model transfer response from the first apparatus 110. The mode transfer response may indicate the subgroup of second configuration identifiers.

[0111] In some example embodiments, the first apparatus 110 receives (2100) a second model from the OTT server 210. In other words, the OTT server 210 transmits (2100) the second model to the first apparatus 110. The second model may be trained at the OTT server based on datasets associated with the subgroup of second configuration identifiers.

[0112] In some example embodiments, the first apparatus 110 transmits (2105) a model transfer request to the second apparatus 120. In other words, the second apparatus 120 receives (2105) the model transfer request from the first apparatus 110. The model transfer request may include a group of first configuration identifiers.

[0113] In some example embodiments, the second apparatus 120 determines (2107) a subgroup of first configuration identifiers from the group of first configuration identifiers. In this case, first training configurations identified by respective the subgroup of first configuration identifiers may be compatible.

[0114] In some example embodiments, the first apparatus 110 receives (2110) a model transfer response from the second apparatus 120. In other words, the second apparatus 120 transmits (2110) the model transfer response to the first apparatus 110. The model transfer response may indicate the subgroup of first configuration identifiers. In some other example embodiments, the first apparatus 110 transmits (2115) a third model to the second apparatus 120. In other words, the second apparatus 120 may receive (2115) the third model from the first apparatus 110. The third model may be trained based on training configurations with the subgroup of first configuration identifiers.

[0115] According to example embodiments describe with reference to FIG. 2A and FIG. 2B, the data samples can be properly selected, which can improve the accuracy of the AI / ML model. It is noted the order of steps shown in FIG. 2 A and FIG. 2B is only an example not limitation.

[0116] Example embodiments of additional conditions grouping are described in detail with reference to FIG. 3 and example embodiments of model training are described in detail with reference to FIG. 4.

[0117] Reference is made to FIG. 3, which illustrates a signaling flow of user equipment side and grouping of additional conditions in accordance with some example embodiments of the present disclosure. For the purpose of discussion, the signaling flow 300 may be implemented in the communication environment 100 shown in FIG. 1. For example, the first apparatus 110 in FIG. 1 may act as the UE. In some embodiments, the second apparatus 120 may act as the network device.

[0118] As the FIG. 3 shows, the second apparatus 120 may define (301) scenario and choose 1 of M “network additional conditions”. The second apparatus 120 may transmit (302), to the first apparatus 110, configure data collection (NW additional conditions encode ID (Am), (E) {Configuration (scenario, RS config, DataMeasConfig, DataLogConfig)}). At these steps, the second apparatus 120 initiates the data collection to the first apparatus 110 by choosing an initial set of “additional conditions” at network side from a sample space of Ito M entries and each vector in this space with an index m and hence the encoded vector Am representing one of those vectors. The network “additional conditions” can indicate for example the gNB antenna panel configuration which may contain network proprietary information such as number of antenna elements and grouping of antenna elements to antenna ports, information about baseband version of the product being used and so on, which is information that cannot be publicly disclosed and hence the encoding. The encoding of the data collection configuration pertaining toa scenario, reference signal config, measurement configuration and logging aspects for the data collection are encoded in the vector E.

[0119] The first apparatus 110 may set (303) scenario and choose N “UE additional conditions”. At this step, the first apparatus 110 sets the reference scenario as the one indicated in Step 302. The novel aspect is that the first apparatus 110 may set some “additional conditions” before initiating the data collection. This is quite proprietary information which cannot be publicly disclosed e.g., number of UE antenna panels and their relative placement, the antenna port as given by its manufacture, orientation of antenna panels, number of UE antenna elements, antenna panel switching duty cycle, number of CPU cores and physical memory assigned for data collection, physical sampling rates, hardware (HW) and software (SW) baseband version of the chipset in use. These are 1 to N entries and the encoding of this results in a vector Cn.

[0120] The first apparatus 110 and the second apparatus 120 may perform (304) data collection initialization handshake. Then the first apparatus 110 and the second apparatus 120 may transmit (305) needed reference signals as per configuration to each other. After the transmission of reference signals, the first apparatus 110 and the second apparatus 120 may perform (306) data collection complete handshake. At these steps, for each vector Cn the data collection proceeds with an initialization handshake at Step 304, transmission of necessary reference signals and alignment of configuration at both the second apparatus 120 and the first apparatus 110 in Step 305, handshake of completion of data collection in Step 306 and the novel aspect in Step 307 of first apparatus 110 indicating the encoded vector Cn to the second apparatus 120 .

[0121] The first apparatus 110 may store (308) UE side data buffer (data samples at NW, Am, Cn), the second apparatus 120 may store (309) NW side data buffer (data samples at UE, Am, Cn). The second apparatus 120 and the first apparatus 110 may have separate data buffers, for each data buffer as a result of an iteration of data collection due to Am and Cn are stored at the second apparatus 120 and the first apparatus 110. The steps 301- 309 may not preclude that data is only collected and stored at one side, namely at the UE or the network. The steps are described to show how data collection is organized at both the sides to explicitly describe the aspects involved at both sides.

[0122] In some example embodiments, the second apparatus 120 transmits (310) NW additional conditions indication (NWAddlCondGrouping). In some other example embodiments, the first apparatus 110 finalizes (311) UE side data buffer (Data Samples per set of {Am, Cn, E}) and UEAddlCondGrouping, NWAddlCondGrouping) toDataSetID UE. In some example embodiments, the first apparatus 110 transmits (312) UE additional conditions indication (UE AddlCondGrouping). In some other example embodiments, the second apparatus 120 finalizes (313) NW side data buffer (Data Samples per set of {Am, Cn, E}) and UEAddlCondGrouping, NW AddlCondGrouping) to DataSetID NW.

[0123] At the step 310 to 313, the second apparatus 120 may signal the NW AddlCondGrouping to indicate to the first apparatus 110 which of the encoded vectors from the network point of view are compatible e.g., antenna panel orientation and number of elements is the same, reference signal configuration and grid of beams enabled are the same. The reason to indicate the group of compatible vectors is the following:• Data collected for each of the compatible vector shares similar characteristics (e.g., sample mean and variance, signal strength values conform to a given range, etc.,);• Model training may mix data from data sets with compatible vectors to improve the robustness during model training;• Model training may mix data from data sets with incompatible vectors to improve model generalization performance;• Trained model should reflect what kind of mixity between the compatible and incompatible vectors have been used in the model training - remember that neighbouring gNB(s) should still be able to use the functionality of the ML model when trained with data sets mixed based on compatible vectors or not use the functionality when vector is not compatible for a given gNB.• Hyperparameters used for model training;• Input data pre-processing;• output data post-processing.

[0124] The first apparatus 110 may signal the UEAddlCondGrouping to the second apparatus 120 which of the encoded vectors from the network point of view are compatible with the criteria on the same lines as described above for NWAddlCondGrouping. Both the second apparatus 120 and the first apparatus 110 mark the final data buffer including of data samples across the combinations of the “additional conditions” as well as marked by UEAddlCondGrouping and NWAddlCondGrouping. The data batch and all its information are pointed by an identity DataSetID UE and DataSetID NW. The expectation at the end of executing the steps in FIG. 3 is that there is data collected at the first apparatus 110 and the second apparatus 120 with an encoding of UE side “additional conditions” and furthermore with an understanding of how these “additional conditions”relate to each other i.e., which ID(s) are compatible or incompatible with each other.

[0125] In some example embodiments, encoding of Am and Cn takes a combination of NW-side additional conditions (antenna element assumption 1, beam codebook assumption 1, etc..) can be refer as one vector with X entries (number of entries depends on the number of different additional conditions. This may often depend on the use-case and in this example beam management has been considered). Per each entry in the vector, entry depend on the number of possibilities for an additional condition. For example, the beam management (BM) codebook may have 12 possibilities while antenna setting may have only 4 possibilities and another additional condition may have 8 possibilities. So, the number of combinations of NW-side additional conditions may be all possible vectors that the second apparatus 120 might have. The space of all the combinations is then M = 12 x 4 x 8 = 384 combinations. The second apparatus 120 may only share the index in this case log2(384) i.e., 9 bits are enough. The second apparatus 120 may add some more bits to not share the index directly but add some kind of encryption and make these 16 bits (the encryption algorithm and the number of bits added may be agreed in the standard or are vendor specific). In this case 16 bits is the encoded ID for Am.

[0126] In some other example embodiments, the UE-side “additional conditions” for the beam management use case may consider (receiver antenna panel element assumption 1, number of antenna panels configured assumption 2, antenna panel switching algorithm assumption 3, physical memory configuration assumption 4, baseband receiver architecture assumption 5, etc..) can be refer as one vector of 5 entries. The number of UE-side “additional conditions” formed based on this may be 14 bits and the first apparatus 110 may add some encryption to make these 24 bits (the encryption algorithm and the number of bits added may be agreed in the standard or are vendor specific). In this case 24 bits is the encoded ID for Cn, as shown in FIG. 4.

[0127] Reference is made to FIG. 4, which illustrates a signaling flow of model training and identification using the information in NW and UE sided additional conditions grouping in accordance with some example embodiments of the present disclosure. For the purpose of discussion, the signaling flow 400 may be implemented in the communication environment 100 shown in FIG. 1. For example, the first apparatus 110 in FIG. 1 may act as the UE. In some embodiments, the second apparatus 120 may act as the network, 0AM, LMF and so on.

[0128] The step 401 is model training for UE at network, 0AM, LMF and so on, mixing data based on NWAddlCondGrouping and UEAddlCondGrouping. The second apparatus120 may transfer (402) ML model request (UEAddlCondGrouping) to the first apparatus 110. After the step 402, the first apparatus 110 may check (403) for compatible vectors in UEAddlCondGrouping. Then the first apparatus 110 may transfer (404) ML model response (compatible vectors) to the second apparatus 120. The first apparatus 110 may continue (405) ML model transfer for compatible vectors. These steps describe when the ML model is being trained at the network side (e.g., gNB, 0AM, LMF, CN, etc.,). As indicated above the ML model can now be trained taking UEAddlCondGrouping and NWAddlCondGrouping information into account to realize different types of ML models, generalized ones which consider mixing data samples across the incompatible vectors or creating a more local ML model by mixing data samples across the compatible vectors or a mix of both. This is proprietary implementation of how the mixing of data samples happens and what is chosen and what is left out etc., but the input is definitely based on the UEAddlCondGrouping and NWAddlCondGrouping. When the network is training such a model for the UE and the ML model has to be downloaded to a new UE, the compatibility check may be requested with the first apparatus 110 to ensure that at least some of the compatible vectors are supported by the first apparatus 110 in the UEAddlCondGrouping.

[0129] The following steps shows the model training at first apparatus 110 or OTT server 420. The step 406 is model training for UE at OTT mixing data based on NWAddlCondGrouping and UEAddlCondGrouping. The OTT server 420 may transfer (407) ML model request (UEAddlCondGrouping) to the first apparatus 110. After the step 407, the first apparatus 110 and the OTT server 420 may check (408) for compatible vectors in UEAddlCondGrouping. Then the first apparatus 110 may transfer (409) ML model response (compatible vectors) to the OTT server 420. The first apparatus 110 and the OTT server 420 may continue (410) ML model transfer for compatible vectors. Then the first apparatus 110, the OTT server 420 and second apparatus 120 may continue (411) ML model identification shown in step 413 onwards. These steps describe the steps when the ML model is being trained at the OTT server 420. The aspects are similar to step 401- 405. Finally, the first apparatus 110 in step 411 may initiate steps 413 onwards to perform ML model identification.

[0130] The following steps shows ML model trained at the first apparatus 110. The step 412 is model training for UE at OTT mixing data based on NWAddlCondGrouping and UEAddlCondGrouping. The first apparatus 110 may transmit (413) ML model identification request (NWAddlCondGrouping) to the second apparatus 120. After thetransmission, the second apparatus 120 may check (414) for compatible vectors in NWAddlCondGrouping. The second apparatus 120 may transmit (415) ML model identification response (compatible vectors) to the first apparatus 110. At step 416, the first apparatus 110, the OTT server 420 and the second apparatus 120 may use only functionality from ML models in compatible vectors. These steps describe when the ML model is being trained at the UE side based on the UEAddlCondGrouping and NWAddlCondGrouping. The mixing guideline may come from OTT server or network may provide this information. After the model is trained the model identification for a UE when arriving at the network may transfer the NWAddlCondGrouping to the network to assess which are the compatible NW additional conditions that the network may have. If there are any compatible vectors the network may continue to use the functionality for those ML models and ignore the rest.

[0131] FIG. 5 shows a flowchart of an example method 500 implemented at a first device in accordance with some example embodiments of the present disclosure. For the purpose of discussion, the method 500 will be described from the perspective of the first apparatus 110 in FIG. 1.

[0132] At block 510, the first apparatus 110 receives, from a second apparatus, model feature information comprising a plurality of first training configurations identified by respective first configuration identifiers, wherein the plurality of first training configurations is associated with communication condition.

[0133] At block 520, the first apparatus 110 determines, for each first training configuration, a plurality of second training configurations identified by respective second configuration identifiers.

[0134] At block 530, the first apparatus 110 obtains, for each second training configuration, a plurality of datasets associated with communication condition based on the plurality of first training configurations, wherein one dataset is obtained based on one first training configuration from the plurality of first training configuration.

[0135] At block 540, the first apparatus 110 transmits, to the second apparatus, data collection information comprising the plurality of second training configurations identified by respective second configuration identifiers.

[0136] In some example embodiments, the method 500 further comprises: storing the plurality of datasets in a buffer which stores data according to a corresponding first training configuration and a corresponding second training configuration.

[0137] In some example embodiments, the method 500 further comprises: receiving,from the second apparatus, a first indication indicating a set of first configuration identifiers, wherein a set of first training configurations identified by respective the set of first configuration identifiers is compatible with each other.

[0138] In some example embodiments, the method 500 further comprises: determining a set of second configuration identifiers, wherein a set of second training configurations identified by respective the set of second configuration identifiers is compatible with each other; determining a set of datasets from the plurality of datasets based on the set of first configuration identifiers and the set of second configuration identifiers; storing the set of datasets together with the set of first configuration identifiers and the set of second configuration identifiers in the buffer; and determining a dataset identity of the set of datasets.

[0139] In some example embodiments, the method 500 further comprises: training a first model based on the set of datasets.

[0140] In some example embodiments, the method 500 further comprises: transmitting, to the second apparatus, a second indication indicating the set of second configuration identifiers.

[0141] In some example embodiments, the method 500 further comprises: receiving, from the second apparatus, a model transfer request, wherein the model transfer request comprising a group of second configuration identifiers; determining a subgroup of second configuration identifiers from the group of second configuration identifiers, wherein second training configurations identified by respective the subgroup of second configuration identifiers are compatible; transmitting, to the second apparatus, a model transfer response indicating the subgroup of second configuration identifiers; and receiving, from the second apparatus, a second model that is trained based on training configurations with the subgroup of second configuration identifiers.

[0142] In some example embodiments, the method 500 further comprises: transmitting, to the second apparatus, a model transfer request, wherein the model transfer request comprising a group of first configuration identifiers; receiving, from the second apparatus, a model transfer response indicating the subgroup of first configuration identifiers; and transmitting, to the second apparatus, a third model that is trained based on training configurations with the subgroup of first configuration identifiers.

[0143] In some example embodiments, the first apparatus comprises a terminal device, and the second apparatus is a core network device or a radio access network device.

[0144] FIG. 6 shows a flowchart of an example method 600 implemented at a seconddevice in accordance with some example embodiments of the present disclosure. For the purpose of discussion, the method 600 will be described from the perspective of the second apparatus 120 in FIG. 1.

[0145] At block 610, the second apparatus 120 transmits, to a first apparatus, model feature information comprising a plurality of first training configurations identified by respective first configuration identifiers, wherein the plurality of first training configurations is associated with communication condition.

[0146] At block 620, the second apparatus 120, for each first training configuration, receives, from the first apparatus, data collection information comprising a plurality of second training configurations identified by respective second configuration identifiers.

[0147] At block 630, the second apparatus 120 obtains, for each second training configuration, a plurality of datasets associated with communication condition based on the plurality of second training configurations, wherein one dataset is obtained based on one second training configuration from the plurality of second training configuration.

[0148] In some example embodiments, the method 600 further comprises: storing the plurality of datasets in a buffer which stores data according to a corresponding first training configuration and a corresponding second training configuration.

[0149] In some example embodiments, the method 600 further comprises: transmitting, to the first apparatus, a first indication indicating a set of first configuration identifiers, wherein characteristic differences among a set of first training configurations identified by respective the set of first configuration identifiers are smaller than a threshold.

[0150] In some example embodiments, the method 600 further comprises: receiving, from the first apparatus, a second indication indicating a set of second configuration identifiers, wherein a set of second training configurations identified by respective the set of second configuration identifiers is compatible with each other.

[0151] In some example embodiments, the method 600 further comprises: determining a set of datasets from the plurality of datasets based on the set of first configuration identifiers and the set of second configuration identifiers; storing the set of datasets together with the set of first configuration identifiers and the set of second configuration identifiers in the buffer; and determining a dataset identity of the set of datasets.

[0152] In some example embodiments, the method 600 further comprises: training a first model based on the set of datasets.

[0153] In some example embodiments, the method 600 further comprises: transmitting, to the first apparatus, a model transfer request, wherein the model transfer requestcomprising a group of second configuration identifiers; receiving, from the first apparatus, a model transfer response indicating the subgroup of second configuration identifiers; and transmitting, to the first apparatus, a second model that is trained based on training configurations with the subgroup of second configuration identifiers.

[0154] In some example embodiments, the method 600 further comprises: receiving, from the first apparatus, a model transfer request, wherein the model transfer request comprising a group of first configuration identifiers; determining a subgroup of first configuration identifiers from the group of first configuration identifiers, wherein first training configurations identified by respective the subgroup of first configuration identifiers are compatible; transmitting, to the first apparatus, a model transfer response indicating the subgroup of second configuration identifiers; and receiving, from the first apparatus, a third model that is transmitted based on training configurations with the subgroup of first configuration identifiers.

[0155] In some example embodiments, the first apparatus comprises a terminal device, and the second apparatus is a core network device or a radio access network device.

[0156] In some example embodiments, a first apparatus capable of performing any of the method 500 (for example, the first apparatus 110 in FIG. 1) may comprise means for performing the respective operations of the method 500. The means may be implemented in any suitable form. For example, the means may be implemented in a circuitry or software module. The first apparatus may be implemented as or included in the first apparatus 110 in FIG. 1.

[0157] In some example embodiments, the first apparatus comprises means for receiving, from a second apparatus, model feature information comprising a plurality of first training configurations identified by respective first configuration identifiers, wherein the plurality of first training configurations is associated with communication condition; means for determining, for each first training configuration, a plurality of second training configurations identified by respective second configuration identifiers; means for obtaining, for each second training configuration, a plurality of datasets associated with communication condition based on the plurality of first training configurations, wherein one dataset is obtained based on one first training configuration from the plurality of first training configuration; and means for transmitting, to the second apparatus, data collection information comprising the plurality of second training configurations identified by respective second configuration identifiers.

[0158] In some example embodiments, the first apparatus further comprises: means forstoring the plurality of datasets in a buffer which stores data according to a corresponding first training configuration and a corresponding second training configuration.

[0159] In some example embodiments, the first apparatus further comprises: means for receiving, from the second apparatus, a first indication indicating a set of first configuration identifiers, wherein a set of first training configurations identified by respective the set of first configuration identifiers is compatible with each other.

[0160] In some example embodiments, the first apparatus further comprises: means for determining a set of second configuration identifiers, wherein a set of second training configurations identified by respective the set of second configuration identifiers is compatible with each other; means for determining a set of datasets from the plurality of datasets based on the set of first configuration identifiers and the set of second configuration identifiers; means for storing the set of datasets together with the set of first configuration identifiers and the set of second configuration identifiers in the buffer; and means for determining a dataset identity of the set of datasets.

[0161] In some example embodiments, the first apparatus further comprises: means for training a first model based on the set of datasets.

[0162] In some example embodiments, the first apparatus further comprises: means for transmitting, to the second apparatus, a second indication indicating the set of second configuration identifiers.

[0163] In some example embodiments, the first apparatus further comprises: means for receiving, from the second apparatus, a model transfer request, wherein the model transfer request comprising a group of second configuration identifiers; means for determining a subgroup of second configuration identifiers from the group of second configuration identifiers, wherein second training configurations identified by respective the subgroup of second configuration identifiers are compatible; means for transmitting, to the second apparatus, a model transfer response indicating the subgroup of second configuration identifiers; and means for receiving, from the second apparatus, a second model that is trained based on training configurations with the subgroup of second configuration identifiers.

[0164] In some example embodiments, the first apparatus further comprises: means for transmitting, to the second apparatus, a model transfer request, wherein the model transfer request comprising a group of first configuration identifiers; means for receiving, from the second apparatus, a model transfer response indicating the subgroup of first configuration identifiers; and means for transmitting, to the second apparatus, a thirdmodel that is trained based on training configurations with the subgroup of first configuration identifiers.

[0165] In some example embodiments, the first apparatus comprises a terminal device, and the second apparatus is a core network device or a radio access network device.

[0166] In some example embodiments, the first apparatus further comprises means for performing other operations in some example embodiments of the method 500 or the first apparatus 110. In some example embodiments, the means comprises at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the performance of the first apparatus.

[0167] In some example embodiments, a second apparatus capable of performing any of the method 600 (for example, the second apparatus 120 in FIG. 1) may comprise means for performing the respective operations of the method 600. The means may be implemented in any suitable form. For example, the means may be implemented in a circuitry or software module. The second apparatus may be implemented as or included in the second apparatus 120 in FIG. 1.

[0168] In some example embodiments, the second apparatus comprises means for transmitting, to a first apparatus, model feature information comprising a plurality of first training configurations identified by respective first configuration identifiers, wherein the plurality of first training configurations is associated with communication condition; means for, for each first training configuration, receiving, from the first apparatus, data collection information comprising a plurality of second training configurations identified by respective second configuration identifiers; and means for obtaining, for each second training configuration, a plurality of datasets associated with communication condition based on the plurality of second training configurations, wherein one dataset is obtained based on one second training configuration from the plurality of second training configuration.

[0169] In some example embodiments, the second apparatus further comprises: means for storing the plurality of datasets in a buffer which stores data according to a corresponding first training configuration and a corresponding second training configuration.

[0170] In some example embodiments, the second apparatus further comprises: means for transmitting, to the first apparatus, a first indication indicating a set of first configuration identifiers, wherein characteristic differences among a set of first training configurations identified by respective the set of first configuration identifiers are smallerthan a threshold.

[0171] In some example embodiments, the second apparatus further comprises: means for receiving, from the first apparatus, a second indication indicating a set of second configuration identifiers, wherein a set of second training configurations identified by respective the set of second configuration identifiers is compatible with each other.

[0172] In some example embodiments, the second apparatus further comprises: means for determining a set of datasets from the plurality of datasets based on the set of first configuration identifiers and the set of second configuration identifiers; means for storing the set of datasets together with the set of first configuration identifiers and the set of second configuration identifiers in the buffer; and means for determining a dataset identity of the set of datasets.

[0173] In some example embodiments, the second apparatus further comprises: means for training a first model based on the set of datasets.

[0174] In some example embodiments, the second apparatus further comprises: means for transmitting, to the first apparatus, a model transfer request, wherein the model transfer request comprising a group of second configuration identifiers; means for receiving, from the first apparatus, a model transfer response indicating the subgroup of second configuration identifiers; and means for transmitting, to the first apparatus, a second model that is trained based on training configurations with the subgroup of second configuration identifiers.

[0175] In some example embodiments, the second apparatus further comprises: means for receiving, from the first apparatus, a model transfer request, wherein the model transfer request comprising a group of first configuration identifiers; means for determining a subgroup of first configuration identifiers from the group of first configuration identifiers, wherein first training configurations identified by respective the subgroup of first configuration identifiers are compatible; means for transmitting, to the first apparatus, a model transfer response indicating the subgroup of second configuration identifiers; and means for receiving, from the first apparatus, a third model that is transmitted based on training configurations with the subgroup of first configuration identifiers.

[0176] In some example embodiments, the first apparatus comprises a terminal device, and the second apparatus is a core network device or a radio access network device.

[0177] In some example embodiments, the second apparatus further comprises means for performing other operations in some example embodiments of the method 600 or thesecond apparatus 120. In some example embodiments, the means comprises at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the performance of the second apparatus.

[0178] FIG. 7 is a simplified block diagram of a device 700 that is suitable for implementing example embodiments of the present disclosure. The device 700 may be provided to implement a communication device, for example, the first apparatus 110 or the second apparatus 120 as shown in FIG. 1. As shown, the device 700 includes one or more processors 710, one or more memories 720 coupled to the processor 710, and one or more communication modules 740 coupled to the processor 710.

[0179] The communication module 740 is for bidirectional communications. The communication module 740 has one or more communication interfaces to facilitate communication with one or more other modules or devices. The communication interfaces may represent any interface that is necessary for communication with other network elements. In some example embodiments, the communication module 740 may include at least one antenna.

[0180] The processor 710 may be of any type suitable to the local technical network and may include one or more of the following: general purpose computers, special purpose computers, microprocessors, digital signal processors (DSPs) and processors based on multicore processor architecture, as non-limiting examples. The device 700 may have multiple processors, such as an application specific integrated circuit chip that is slaved in time to a clock which synchronizes the main processor.

[0181] The memory 720 may include one or more non-volatile memories and one or more volatile memories. Examples of the non-volatile memories include, but are not limited to, a Read Only Memory (ROM) 724, an electrically programmable read only memory (EPROM), a flash memory, a hard disk, a compact disc (CD), a digital video disk (DVD), an optical disk, a laser disk, and other magnetic storage and / or optical storage. Examples of the volatile memories include, but are not limited to, a random access memory (RAM) 722 and other volatile memories that will not last in the power-down duration.

[0182] A computer program 730 includes computer executable instructions that are executed by the associated processor 710. The instructions of the program 730 may include instructions for performing operations / acts of some example embodiments of the present disclosure. The program 730 may be stored in the memory, e.g., the ROM 724. The processor 710 may perform any suitable actions and processing by loading theprogram 730 into the RAM 722.

[0183] The example embodiments of the present disclosure may be implemented by means of the program 730 so that the device 700 may perform any process of the disclosure as discussed with reference to FIG. 2 to FIG. 6. The example embodiments of the present disclosure may also be implemented by hardware or by a combination of software and hardware.

[0184] In some example embodiments, the program 730 may be tangibly contained in a computer readable medium which may be included in the device 700 (such as in the memory 720) or other storage devices that are accessible by the device 700. The device 700 may load the program 730 from the computer readable medium to the RAM 722 for execution. In some example embodiments, the computer readable medium may include any types of non-transitory storage medium, such as ROM, EPROM, a flash memory, a hard disk, CD, DVD, and the like. The term “non-transitory,” as used herein, is a limitation of the medium itself (i.e., tangible, not a signal) as opposed to a limitation on data storage persistency (e.g., RAM vs. ROM).

[0185] Generally, various embodiments of the present disclosure may be implemented in hardware or special purpose circuits, software, logic or any combination thereof. Some aspects may be implemented in hardware, and other aspects may be implemented in firmware or software which may be executed by a controller, microprocessor or other computing device. Although various aspects of embodiments of the present disclosure are illustrated and described as block diagrams, flowcharts, or using some other pictorial representations, it is to be understood that the block, apparatus, system, technique or method described herein may be implemented in, as non-limiting examples, hardware, software, firmware, special purpose circuits or logic, general purpose hardware or controller or other computing devices, or some combination thereof.

[0186] Some example embodiments of the present 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 computerexecutable instructions, such as those included in program modules, being executed in a device on a target physical or virtual processor, to carry out any of the methods as described above. Generally, program modules include routines, programs, libraries, objects, classes, components, data structures, or the like that perform particular tasks or implement particular abstract data types. The functionality of the program modules may be combined or split between program modules as desired in various embodiments.Machine-executable instructions for program modules may be executed within a local or distributed device. In a distributed device, program modules may be located in both local and remote storage media.

[0187] Program code for carrying out methods of the present 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 the program code, when executed by the processor or controller, cause the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may execute entirely on a machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.

[0188] In the context of the present disclosure, the computer program code or related data may be carried by any suitable carrier to enable the device, apparatus or processor to perform various processes and operations as described above. Examples of the carrier include a signal, computer readable medium, and the like.

[0189] The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable medium may include but not limited to an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the computer readable storage medium would include an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0190] Further, although operations are depicted in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed, to achieve desirable results. In certain circumstances, multitasking and parallel processing may be advantageous. Likewise, although several specific implementation details are contained in the above discussions, these should not be construed as limitations on the scope of the present disclosure, but rather as descriptions of features that may be specific to particular embodiments. Unless explicitly stated, certain features that are described in the context of separate embodiments may also be implemented in combination in a single embodiment.Conversely, unless explicitly stated, various features that are described in the context of a single embodiment may also be implemented in a plurality of embodiments separately or in any suitable sub-combination.

[0191] Although the present disclosure has been described in languages specific to structural features and / or methodological acts, it is to be understood that the present disclosure defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the claims.

Claims

I / We Claim:

1. A first apparatus comprising: at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the first apparatus to: receive, from a second apparatus, model feature information comprising a plurality of first training configurations identified by respective first configuration identifiers, wherein the plurality of first training configurations is associated with communication condition; determine, for each first training configuration, a plurality of second training configurations identified by respective second configuration identifiers; obtain, for each second training configuration, a plurality of datasets associated with communication condition based on the plurality of first training configurations, wherein one dataset is obtained based on one first training configuration from the plurality of first training configuration; and transmit, to the second apparatus, data collection information comprising the plurality of second training configurations identified by respective second configuration identifiers.

2. The first apparatus of claim 1, wherein the first apparatus is caused to: store the plurality of datasets in a buffer which stores data according to a corresponding first training configuration and a corresponding second training configuration.

3. The first apparatus of claim 1 or 2, wherein the first apparatus is caused to: receive, from the second apparatus, a first indication indicating a set of first configuration identifiers, wherein characteristic differences among a set of first training configurations identified by respective the set of first configuration identifiers are smaller than a threshold.

4. The first apparatus of claim 3, wherein the first apparatus is caused to: determine a set of second configuration identifiers, wherein a set of second training configurations identified by respective the set of second configuration identifiers is compatible with each other; determine a set of datasets from the plurality of datasets based on the set of first configuration identifiers and the set of second configuration identifiers;store the set of datasets together with the set of first configuration identifiers and the set of second configuration identifiers in the buffer; and determine a dataset identity of the set of datasets.

5. The first apparatus of claim 4, wherein the first apparatus is caused to: train a first model based on the set of first training configurations.

6. The first apparatus of claim 5, wherein the first apparatus is caused to: transmit, to the second apparatus, a second indication indicating the set of second configuration identifiers.

7. The first apparatus of any of claims 1-6, wherein the first apparatus is caused to: receive, from the second apparatus, a model transfer request, wherein the model transfer request comprising a group of second configuration identifiers; determine a subgroup of second configuration identifiers from the group of second configuration identifiers, wherein second training configurations identified by respective the subgroup of second configuration identifiers are compatible; transmit, to the second apparatus, a mode transfer response indicating the subgroup of second configuration identifiers; and receive, from the second apparatus, a second model that is trained based on training configurations associated with the subgroup of second configuration identifiers.

8. The first apparatus of any of claims 1-6, wherein the first apparatus is caused to: transmit, to the second apparatus, a model transfer request, wherein the model transfer request comprising a group of first configuration identifiers; receive, from the second apparatus, a mode transfer response indicating the subgroup of first configuration identifiers; and transmit, to the second apparatus, a third model that is trained based on training configurations associated with the subgroup of first configuration identifiers.

9. The first apparatus of any of claims 1-8, wherein the first apparatus comprises a terminal device, and the second apparatus is a core network device or a radio access network device.

10. A second apparatus comprising: at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the second apparatus to: transmit, to a first apparatus, model feature information comprising a plurality of first training configurations identified by respective first configuration identifiers, wherein the plurality of first training configurations is associated with communication condition; for each first training configuration, receive, from the first apparatus, data collection information comprising a plurality of second training configurations identified by respective second configuration identifiers; and obtain, for each second training configuration, a plurality of datasets associated with communication condition based on the plurality of second training configurations, wherein one dataset is obtained based on one second training configuration from the plurality of second training configuration.

11. The second apparatus of claim 10, wherein the second apparatus is caused to: store the plurality of datasets in a buffer which stores data according to a corresponding first training configuration and a corresponding second training configuration.

12. The second apparatus of claim 10 or 11, wherein the second apparatus is caused to: transmit, to the first apparatus, a first indication indicating a set of first configuration identifiers, wherein characteristic differences among a set of first training configurations identified by respective the set of first configuration identifiers are smaller than a threshold.

13. The second apparatus of claim 12, wherein the second apparatus is caused to: receive, from the first apparatus, a second indication indicating a set of second configuration identifiers, wherein a set of second training configurations identified by respective the set of second configuration identifiers is compatible with each other.

14. The second apparatus of claim 13, wherein the second apparatus is caused to: determine a set of datasets from the plurality of datasets based on the set of first configuration identifiers and the set of second configuration identifiers; store the set of datasets together with the set of first configuration identifiers and the set of second configuration identifiers in the buffer; anddetermine a dataset identity of the set of datasets.

15. The second apparatus of claim 14, wherein the second apparatus is caused to: train a first model based on the set of second training configurations.

16. The second apparatus of any of claims 10-15, wherein the second apparatus is caused to: transmit, to the first apparatus, a model transfer request, wherein the model transfer request comprising a group of second configuration identifiers; receive, from the first apparatus, a model transfer response indicating the subgroup of second configuration identifiers; and transmit, to the first apparatus, a second model that is trained based on training configurations associated with the subgroup of second configuration identifiers.

17. The second apparatus of any of claims 10-15, wherein the second apparatus is caused to: receive, from the first apparatus, a model transfer request, wherein the model transfer request comprising a group of first configuration identifiers; determine a subgroup of first configuration identifiers from the group of first configuration identifiers, wherein first training configurations identified by respective the subgroup of first configuration identifiers are compatible; transmit, to the first apparatus, a mode transfer response indicating the subgroup of second configuration identifiers; and receive, from the first apparatus, a third model that is transmitted based on training configurations associated with the subgroup of first configuration identifiers.

18. The second apparatus of any of claims 10-17, wherein the first apparatus comprises a terminal device, and the second apparatus is a core network device or a radio access network device.

19. A method implemented at a first apparatus, comprising: receiving, from a second apparatus, model feature information comprising a plurality of first training configurations identified by respective first configuration identifiers, wherein the plurality of first training configurations is associated with communication condition;determining, for each first training configuration, a plurality of second training configurations identified by respective second configuration identifiers; obtaining, for each second training configuration, a plurality of datasets associated with communication condition based on the plurality of first training configurations, wherein one dataset is obtained based on one first training configuration from the plurality of first training configuration; and transmitting, to the second apparatus, data collection information comprising the plurality of second training configurations identified by respective second configuration identifiers.

20. A method implemented at a second apparatus, comprising: transmitting, to a first apparatus, model feature information comprising a plurality of first training configurations identified by respective first configuration identifiers, wherein the plurality of first training configurations is associated with communication condition; for each first training configuration, receiving, from the first apparatus, data collection information comprising a plurality of second training configurations identified by respective second configuration identifiers; and obtaining, for each second training configuration, a plurality of datasets associated with communication condition based on the plurality of second training configurations, wherein one dataset is obtained based on one second training configuration from the plurality of second training configuration.

21. A first apparatus comprising: means for receiving, from a second apparatus, model feature information comprising a plurality of first training configurations identified by respective first configuration identifiers, wherein the plurality of first training configurations is associated with communication condition; means for determining, for each first training configuration, a plurality of second training configurations identified by respective second configuration identifiers; means for obtaining, for each second training configuration, a plurality of datasets associated with communication condition based on the plurality of first training configurations, wherein one dataset is obtained based on one first training configuration from the plurality of first training configuration; andmeans for transmitting, to the second apparatus, data collection information comprising the plurality of second training configurations identified by respective second configuration identifiers.

22. A second apparatus comprising: means for transmitting, to a first apparatus, model feature information comprising a plurality of first training configurations identified by respective first configuration identifiers, wherein the plurality of first training configurations is associated with communication condition; means for: for each first training configuration, receiving, from the first apparatus, data collection information comprising a plurality of second training configurations identified by respective second configuration identifiers; and means for obtaining, for each second training configuration, a plurality of datasets associated with communication condition based on the plurality of second training configurations, wherein one dataset is obtained based on one second training configuration from the plurality of second training configuration.

23. A computer readable medium comprising instructions stored thereon for causing an apparatus at least to perform the method of claim 19 or claim 20.

Images