Data collection for an ai / ml model

Abstract

A method comprising: obtain, at a first apparatus by measuring a set of reference signals from a second apparatus, a first set of measurement results to be used to a model; receiving, from the second apparatus, at least one of: first assistance information indicating a first characteristic of a second set of measurement results associated with the set of reference signals, or an indication of a usability of the first set of measurement results as input to the model; and performing an action associated with the first set of measurement results on the model based on the at least one of the first assistance information or the indication

Description

DATA COLLECTION FOR AN AI / ML MODELRELATED APPLICATION

[0001] The present application claims priority from, and the benefit of, US Provisional Application 63 / 518774, filed August 10th, 2023, 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 data collection for an artificial intelligence (AI) / machine learning (ML) model.BACKGROUND

[0003] Several technologies have been proposed to improve communication performances. For example, communication devices may employ an AI / ML model to improve communication qualities. The Al / ML model can be applied to different scenarios. Data collection is needed, for example, to train, re-train, fine-tune the AI / ML model.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: obtain, by measuring a set of reference signals from a second apparatus, a first set of measurement results to be used to a model; receive, from the second apparatus, at least one of: first assistance information indicating a first characteristic of a second set of measurement results associated with the set of reference signals, or an indication of a usability of the first set of measurement results as input to the model; and perform an action associated with the first set of measurement results on the model based on the at least one of the first assistance information or the indication.

[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, a set of reference signals for obtaining a first set of measurement results to be used to a model; and transmit, to the first apparatus, at least one of: first assistance information indicating a first characteristic of a second set of measurement results associated with the set of reference signals, or an indication of a usability of the first set of measurement results as input to the model.

[0006] In a third aspect of the present disclosure, there is provided a third apparatus. The third apparatus comprises at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the third apparatus to: receive, from a first apparatus, a first set of measurement results on a set of reference signals from a second apparatus and first assistance information indicating a first characteristic of a second set of measurement results associated with the set of reference signals; determine a data type of the first set of measurement results based on the first assistance information; and update a model corresponding to the data type of the first set of measurement results based on the first set of measurement results.

[0007] In a fourth aspect of the present disclosure, there is provided a method. The method comprises: obtaining, at a first apparatus by measuring a set of reference signals from a second apparatus, a first set of measurement results to be used to a model; receiving, from the second apparatus, at least one of: first assistance information indicating a first characteristic of a second set of measurement results associated with the set of reference signals, or an indication of a usability of the first set of measurement results as input to the model; and performing an action associated with the first set of measurement results on the model based on the at least one of the first assistance information or the indication.

[0008] In a fifth aspect of the present disclosure, there is provided a method. The method comprises: transmitting, at a second apparatus to a first apparatus, a set of reference signals for obtaining a first set of measurement results to be used to a model; and transmitting, to the first apparatus, at least one of: first assistance information indicating a first characteristic of a second set of measurement results associated with the set of reference signals, or an indication of a usability of the first set of measurement results as input to the model.

[0009] In a sixth aspect of the present disclosure, there is provided a method. Themethod comprises: receiving, at a third apparatus from a first apparatus, a first set of measurement results on a set of reference signals from a second apparatus and first assistance information indicating a first characteristic of a second set of measurement results associated with the set of reference signals; determining a data type of the first set of measurement results based on the first assistance information; and updating a model corresponding to the data type of the first set of measurement results based on the first set of measurement results.

[0010] In a seventh aspect of the present disclosure, there is provided a first apparatus. The first apparatus comprises means for obtaining, by measuring a set of reference signals from a second apparatus, a first set of measurement results to be used to a model; means for receiving, from the second apparatus, at least one of: first assistance information indicating a first characteristic of a second set of measurement results associated with the set of reference signals, or an indication of a usability of the first set of measurement results as input to the model; and means for performing an action associated with the first set of measurement results on the model based on the at least one of the first assistance information or the indication.

[0011] In an eighth aspect of the present disclosure, there is provided a second apparatus. The second apparatus comprises means for transmitting, to a first apparatus, a set of reference signals for obtaining a first set of measurement results to be used to a model; and means for transmitting, to the first apparatus, at least one of: first assistance information indicating a first characteristic of a second set of measurement results associated with the set of reference signals, or an indication of a usability of the first set of measurement results as input to the model.

[0012] In a ninth aspect of the present disclosure, there is provided a third apparatus. The third apparatus comprises means for receiving, from a first apparatus, a first set of measurement results on a set of reference signals from a second apparatus and first assistance information indicating a first characteristic of a second set of measurement results associated with the set of reference signals; means for determining a data type of the first set of measurement results based on the first assistance information; and means for updating a model corresponding to the data type of the first set of measurement results based on the first set of measurement results.

[0013] In a tenth aspect of the present disclosure, there is provided a computer readablemedium. The computer readable medium comprises instructions stored thereon for causing an apparatus to perform at least the method according to the fourth aspect.

[0014] In an eleventh 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 fifth aspect.

[0015] In a twelfth 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 sixth aspect.

[0016] 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

[0017] Some example embodiments will now be described with reference to the accompanying drawings, where:

[0018] FIG. 1 illustrates an example communication environment in which example embodiments of the present disclosure can be implemented;

[0019] FIG. 2 illustrates an example data life cycle for a terminal device-side model training for beam prediction model according to some example embodiments of the present disclosure;

[0020] FIG. 3A and FIG. 3B illustrate statistical measures for all transmitting beams with different antenna array configurations of a network device;

[0021] FIG. 4 illustrates an example architecture of assisted data collection for model update at user equipment (UE) / UE server according to some example embodiments of the present disclosure;

[0022] FIG. 5 illustrates an example signaling flow for determining assistance information according to some example embodiments of the present disclosure;

[0023] FIG. 6 illustrates an example signaling flow for determining assistance information according to some example embodiments of the present disclosure;

[0024] FIG. 7 illustrates an example signaling flow for determining assistance information according to some example embodiments of the present disclosure;

[0025] FIG. 8 illustrates an example architecture of the use of assistance information for data collection according to some example embodiments of the present disclosure;

[0026] FIG. 9 illustrates an example signaling flow for data collection for AI / ML model according to some example embodiments of the present disclosure;

[0027] FIG. 10 illustrates a flowchart of a method implemented at a first apparatus according to some example embodiments of the present disclosure;

[0028] FIG. 11 illustrates a flowchart of a method implemented at a second apparatus according to some example embodiments of the present disclosure;

[0029] FIG. 12 illustrates a flowchart of a method implemented at a third apparatus according to some example embodiments of the present disclosure;

[0030] FIG. 13 illustrates a simplified block diagram of a device that is suitable for implementing example embodiments of the present disclosure; and

[0031] FIG. 14 illustrates a block diagram of an example computer readable medium in accordance with some example embodiments of the present disclosure.

[0032] Throughout the drawings, the same or similar reference numerals represent the same or similar element.DETAILED DESCRIPTION

[0033] 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.

[0034] 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.

[0035] 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.

[0036] 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.

[0037] 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.

[0038] 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.

[0039] 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., but do not preclude the presence or addition of one or more other features, elements, components and / or combinations thereof.

[0040] 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.

[0041] 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.

[0042] 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) 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. Embodimentsof 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.

[0043] 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), 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.

[0044] 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-mounted display (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 devicesoperating 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.

[0045] 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.

[0046] As used herein, the term “AI / ML model” may refer to a data driven algorithm that applies AI / ML techniques to generate a set of outputs based on a set of inputs. In the context of the present disclosure, the term “AI / ML model” may be used interchangeably with the terms “model”, “Al model” and “ML model”. The term “AI / ML” may be used interchangeably with the terms “Al” and “ML”.

[0047] As used herein, the term “characteristic of measurement results” may include any suitable information which can characterizing the measurement results. For example, the characteristic of the measurement results may be related to the configuration for obtaining the measurement results, for example, the network configuration for transmitting reference signals. Alternatively, or in addition, the characteristic of the measurement results may include the statistical properties or data distribution properties of the measurement results, for example, the mean and / or standard deviation, as described below.Example Environment

[0048] FIG. 1 illustrates an example communication environment 100 in which exampleembodiments of the present disclosure can be implemented. As shown in FIG. 1, the communication environment 100 may at least include a first apparatus 110 and a second apparatus 120. The first apparatus 110 may communicate with the second apparatus 120. An AI / ML model (shorted as model hereinafter) may be deployed at the first apparatus 110. The AI / ML model may be used for any suitable use cases or to implement any suitable functionalities, for example but not limited to, channel state information (CSI) prediction, beam management (BM), positioning, etc. In order to update the AL / ML model, data collection may be needed. In the present disclosure, updating the AI / ML model or updating the model or the like may refer to make a change to the model by any suitable means. For example, updating the model may include but not limited to retraining, fine-tuning the model. In the following, the terms “updating”, “fine-tuning” and “retraining” may be used interchangeably.

[0049] It is to be understood that the number of second apparatus and first apparatus shown in FIG. 1 is given for the purpose of illustration without suggesting any limitations. The communication network 100 may include any suitable number of second apparatus and first apparatus.

[0050] In some example embodiments, the first apparatus 110 may comprise a terminal device (for example, a UE), and the second apparatus 120 may comprise a network device (for example, a gNB). If the first apparatus 110 is a UE, the model at the UE may be also referred to as a UE side model.

[0051] In the following, for the purpose of illustration, some example embodiments are described with the first apparatus 110 operating as a terminal device (for example, UE) and the second apparatus 120 operating as a network device (for example, a gNB). 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.

[0052] In some example embodiments, the communication environment 100 may include a third apparatus 130, which is at least in communication with the first apparatus 110. In some example embodiments, if the first apparatus 110 is a UE, the third apparatus 130 may be a UE server or a third-party neutral server.

[0053] Communications in the communication environment 100 may be implementedaccording 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.

[0054] To support a new AI / ME-enabled radio interface for the next cellular systems, 3rd Generation Partnership Project (3GPP) is currently investigating a new study item for Release (Rel) 18. Beam management is one of the different pilot use cases for applying AI / ME model in the NR radio interface. The different pilot use cases will also serve to identify a common AI / ME framework, including functional requirements of AI / ML architecture, which could be used in subsequent releases and will pave the way for the design of native AI / ML 6G networks. In the following, some aspects or example embodiments may be described by taking the use case of beam management as an example.

[0055] The Rel- 18 beam management use case is further studied for spatial and / or time beam prediction. The scope of spatial beam prediction (denoted as BM-Casel) is to predict the best DL Tx beam and / or DL Tx / Rx beam pairs in different spatial locations. Conversely, time-domain beam predictions (denoted as BM-Case2) aim to predict the best DL Tx beam and / or DL Tx / Rx beam pairs beam to use for next time instants. The primary motivation for supporting AI / ML-based beam management is overhead savings and latency reduction. It has been shown that ML algorithms enable predicting the serving beam for different UE locations and time instances, thus avoiding measuring the actual beam quality and saving those resources to be employed for data transmission and / or for the UE to increase the length of DRX. On the other hand, beam scanning operations such as those performed in procedure 1 (Pl), procedure 2 (P2) and procedure 3 (P3) are time inefficient and not scalable when the size of antenna arrays increases. Therefore, MLalgorithms can replace sequential beam scanning by recommending a reduced set of beams likely to contain the best beam index of the full beam scan.

[0056] It is noted that a set of beams to predict are indicated as “Set A” beams. The beam prediction model may predict the best beam ID and / or the RSRP value in Set A. Also, Set A may include Set B (when Set B is subset of Set A) or Set B may be different from Set A, e.g., Set B may be wide beams and Set A may be narrow beams.

[0057] For data collection for training beam prediction model, one of the main aspects currently discussed is data collection for training the AI / ML beam prediction model. Among the different aspects, the signaling aspect, the configuration of measurement and measurements report for data collection are considered. The use of assistance information is also included in the signaling aspect. Other topics include the content and type of the collected data.

[0058] Regarding data collection for gNB-side AI / ML model, one important aspect considers the measurements reporting overhead for data collection. An example option is to report N layer 1 (LI preference signal received powers (RSRPs) (corresponding to the N strongest beams) with the indications / index of the beams. The reporting may be based on the measurement corresponding to a beam set and N may be extended to report a number of beams larger than 4. Solutions that consider quantizing LI -RSRP or normalizing the LI -RSRP measurement with a lower number of bits than the current specification is also under discussion. A second important aspect considers the quality of the data, which includes the reduction of the Ll-RSRP measurement errors introduced due to UE measurement inaccuracies. In some reporting mechanisms, the UE may omit reporting data measured with low quality for reducing the reporting overhead.

[0059] Regarding the data collection for AI / ML model training at the UE side, some mechanisms focus on the potential specification impact of aspects such as data collection triggered by UE, the UE reporting the supported / preferred configurations related to Set A and / or Set B beam measurements and the transmission of assistance information from the network to UE. Regarding data collection for UE-side model training, UE may require flexibility to request the data collection. Channel State Information reference signal (CSI- RS) measurement enhancements may also be required for data collection at the UE side. For instance, by allowing the measurements of full or partial Set A beam (associated with a functionality) to have a longer periodicity than the Set B measurements. For the UE-side model, assistance information may be transmitted from gNB to UE to support the UE training of the model. One example is that the UE may have more details about the relation between Set A beams and Set B beams. A key aspect is about preserving sensitive and proprietary information during the transmission of assistance information.

[0060] In some solutions, regarding data collection for BM-Casel and BM-Case2 with a UE-side AI / ML model, assistance information from gNB to UE for UE data collection may be used for categorizing the data for the purpose of differentiating characteristics of data. The assistance information should preserve privacy / proprietary information.

[0061] The use of the assistance information has been already discussed. Concerning the transmission of assistance information from gNB to UE for UE-side model, information regarding gNB-side beam shape, for example, beam beamwidth, beam boresight directions, and Tx beam angle are sensitive / proprietary information.

[0062] Other information such as the Rel-17 beam antenna information used by the location server to provide relative beam information of the transmission and reception point (TRP) can be considered in a separate discussion. Although Rel-17 beam antenna information can be extremely valuable for assisting the UE model, its use may not be reliable or guaranteed because the presence of Rel-17 beam antenna information is optionally provided by a network node.

[0063] In this way, regarding the explicit assistance information from network to UE for UE-side AI / ML model, there is no consensus to support the following information, including gNB-side beam shape information, such as, 3dB beamwidth, beam boresight directions, beam shape, Tx beam angle, etc. It is noted that other information (e.g., relative information) of Tx beam(s) preserving sensitive proprietary information is a separate discussion.

[0064] For data collection for functionality-based and model-ID-based Life cycle management (LCM), functionality identification and model identification are proposed to define the basic framework for the 3GPP framework for AI / ML. For UE-sided models, as one option, the conditions for supported functionality / functionalities of a given sub-use case (ML-enabled feature) can be identified. In functionality identification and functionality-based LCM, knowing the UE conditions (including parameters / configurations) is required at the network as the first step before any other, as this shall reveal the background conditions when using ML models for supporting a givenML-enabled feature.

[0065] Data collection can be considered as conditions for functionalities associated with both BM-Casel and BM-Case2. A set of conditions for data collection functionality may be applicable for the BM-Casel-2 assuming DL Tx beam prediction.

[0066] The data collection may mainly consider offline training. To better explain important aspect, an example data life cycle 200 for UE-side training of the beam prediction model is shown in FIG. 2. Multiple UEs 221 and 231 may collect data based on the Ll-RSRP measurements corresponding to different beams. The UE is assumed may not be capable of training a beam prediction model from scratch on its own due to i) limited data available at the single UE and ii) hardware limitations such as computational resources and UE power limitation problems. Therefore, the UE may send the beam measurements to an external UE server 210, for instance, located in a datacenter. The UE server 210 may receive data from multiple UEs from the same cell (e.g., UEs 211 served by gNB 220) or from a different cell (e.g., UEs 231 served by gNB 230). The UE server 210 may aggregate the data received from multiple UEs to form a training dataset 211. With a larger dataset, the UE server 210 may be able to train a more general model 240 that can be applied to multiple site locations. The data may be pre-processed using normalization, scaling, etc., before model training. Lastly, the beam prediction model is downloaded to all the UEs.

[0067] During the UE connection, a UE may require updating / fine-tuning the beam prediction model for several reasons. For instance, the model 240 may require some updates due to the UE configuration, or due the channel propagation conditions between gNB and UE, or any other reasons. Model update / fine-tuning may be performed at the UE since i) it is more effective by using the dataset collected locally at the UE and ii) it may not involve large computational resources overhead. Alternatively, and / or additionally, the UE may send the beam measurement data to the UE server 210, in which the same sequence of steps described for model training may be also applied for updating the model 240 with data from multiple UEs.

[0068] One fundamental issue for UE-side model training is that the UE collecting data may not be aware of the type of data that is collecting. Consequently, the UE (including a UE server) may use incorrect data for model updates, or the UE may update the wrong model, i.e., a model trained with other type of data.

[0069] In one example, the data used for the model update are measured based on a different antenna array configuration than that used by the gNB during model training. Hence, the model update may not converge because the model parameters were trained using a different data distribution. For example, the average RSRP and standard deviation of the RSRP values measured over the entire dataset is considered, which consists of multiple Tx beam measurements (performed for both the serving and non-serving beam) over several hundreds of UEs. The FIG. 3A show the average RSRPs for all Tx beams with three datasets 311, 312, 313 generated with a gNB antenna array consisting of 2x4 antenna elements, 4x8 antenna elements and 8x16 antenna elements. The FIG. 3B shows the standard deviation of the RSRPs for all Tx beams with three datasets generated with a gNB antenna array consisting of 2x4 antenna elements 321, 4x8 antenna elements 322 and 8x16 antenna elements 323, respectively. Adopting a larger gNB antenna array configuration, i.e., increasing the number of antenna elements, causes the gNB to form narrower beams, which concentrate the transmitted power in a smaller region than wide beams. Therefore, the mean RSRP becomes lower for the narrower beams due to the average computed for both serving and non-serving beams.

[0070] Mean and standard deviation of RSRP measurements may be used in the data pre-processing step for scaling the RSRP measurements within a range of values more adequate to be used as input for ML models. Consequently, the mean RSRP used by the standard scaler is lower with the larger gNB antenna array, meaning that the mean RSRP used by the standard scaler should be updated accordingly to the configurations associated with the data sample. These additional pre-processing steps may be more practical if the UE is able to classify data belonging to a specific configuration.

[0071] In another example, the UE server may collect a dataset with a gNB antenna array configuration characterized by mean pi and standard deviation oi and applies a standard scaler during the model training. Later, a UE may decide to update the model with the beam measurements collected locally. One or more conditions like the gNB antenna array configuration may be changed (e.g., due to changing on the gNB codebook) compared to the ones applied during model training. Therefore, the new set of measurements is characterized by mean 2^pi and standard deviation 02^01, which differs from the trained model. The UE would like to switch to a different model that is trained with the set of measurements characterized by mean 2 and standard deviation 02. Therefore, the UE may proceed to download the model from the UE server and update itby using mean p2 and standard deviation 02 applied to the standard scaler.

[0072] From these examples, the UE knowledge about the type of data collected becomes key for UE side data collection and correctly operating the model. The use of assistance information may be a solution. Nevertheless, a critical aspect tackled in the present disclosure is to preserve gNB-side proprietary information like gNB beam shape, Tx beam angle, beam boresight direction, etc., while providing enough information to help the UE categorizing the data collected.Work Principle and Example architecture

[0073] According to some example embodiments of the present disclosure, there is provided a solution for data collection for AI / ML model. According to the example embodiments, assistance information indicating a characteristic of measurement results may be transferred between a first apparatus (for example, UE) and a second apparatus (for example, gNB). In some example embodiments, first assistance information may be transmitted from the second apparatus and the first apparatus may determine whether to update a model or switch to another model based on the first assistance information. In the following, such example embodiments may be referred to as option A. In some example embodiments, second assistance information may be transmitted from the first apparatus to the second apparatus. The second apparatus may determine an action to be performed on the model based on the second assistance information and indicate the determined action to the first apparatus, for example, whether to update the ML or switch to another model. In the following, such example embodiments may be referred to as option B.

[0074] The proposed solution may be employed for any AI / ML use case or functionality, for example, but not limited to, beam prediction or beam management, CSI prediction, positioning.

[0075] To better understand the proposed solution of the present disclosure, some example embodiments are first described with respect the use case of beam prediction. In this use case, the measurement result may be the RSRPs, also referred to as Ll-RSRP measurements. Moreover, the UE is described as an example of the first apparatus 110 and the gNB is described as an example of the second apparatus 120, and the UE server is described as an example of the third apparatus 130.

[0076] FIG. 4 illustrates an example architecture 400 of assisted data collection for model update at UE / UE server according to some example embodiments of the present disclosure. For the ML-based beam prediction by using the UE-side model, it is possible to use assistance information to help the UE 421 to classify the type of beam measurements collected and use correct type of data for model updates. The use of assistance information, like the mean RSRP and standard deviation of the RSRP values measured over the entire dataset 411, preserve gNB-side proprietary information and does not reveal the gNB 420 Tx beam codebook design.

[0077] The example steps are explained with the steps below. At step 1), the UE 421 may transmit a message requesting the data collection configuration for updating UE-side model 412. The message may include an explicit indication requesting the gNB 420 to initiate transmission of the signaling needed for data collection. The gNB 420 may acknowledge the request or post-postpone it until the gNB 420’ s conditions are favorable for starting a new data collection phase. The gNB 420 may respond to the data collection request message by transmitting the CSI-RS associated with the beams in Set A / Set B and allow the UE 421 to collect new beam measurements for training / updating / fine-tuning the UE-side model. The message may be included as part of the applicable functionality framework which enables the UE 421 to report applicable functionalities as well as additional conditions to each functionality (i.e., the gNB 420 configured functionalities).

[0078] At step 2), the UE 421 may transmit, to the gNB 420, a message indicating the need for assistance information for data collection assistance. The UE 421 may indicate the need for assistance information with respect to the supported functionalities. As an example, the assistance information may be indicated as “to be reported from gNB”. One or more conditions may indicate the need of assistance information required at the UE 421 for one or more functions, like training, inference, performance monitoring. The gNB 420 may determine and transmit the assistance information only when one or more need for assistance information transmission is indicated “to be reported” in the message from the UE 421. In some example embodiment, the gNB 420 may configure the UE 421 to report assistance information associated with data collection.

[0079] At step 3), the gNB 420 may determine the assistance information for data collection. The gNB-side proprietary information is preserved. For example, the assistance information may be determined based on the mean and standard deviation ofthe Ll-RSRP measurements associated with the current network (NW) configuration used by the gNB 420 for CSI-RS transmission. In one example, the mean and standard deviation of the Ll-RSRP measurements may be determined per NW configuration. The gNB 420 may repeat the determination of the assistance information at every change of the NW configuration. In some example embodiment, the mean and standard deviation of the Ll- RSRP measurements may be determined per Tx beam, i.e., for each model input feature.

[0080] Alternatively, in some example embodiments, the UE 421 may determine the assistance information for the current model deployed at UE 421. In an example, the UE 421 may receive from the UE server 410 the standard scaler parameters, i.e., the mean and standard deviation of Ll-RSRP associated with the UE 421 side model and used in the pre-processing steps while training the UE-side model 410. In another example, the UE 421 may determine the mean and standard deviation of Ll-RSRP beam measurements used if training the model on its own.

[0081] At step 4), the gNB 420 may transmit to the UE 421 a message including assistance information to support the UE data collection. In some example embodiment, the assistance information for UE data collection may be initially configured from gNB420 (e.g., via RRC). In such a way, the UE 421 may use the assistance information, while performing beam measurements. In addition, a dynamic indication of the use of assistance information for UE data collection may be possible with an MAC CE. In one example, one or more messages may be sent via the MAC CE in case the gNB 420 dynamically changes the configuration. A first MAC CE message may indicate to update or enable / disable the use of assistance information. The triggering of data collection for the new network configuration may be indicated by the first MAC CE message or alternatively may be indicated in a second MAC CE message. In another example, the UE421 may report the assistance information associated with the UE-side model 410 such as the data characteristics, for example, the mean and standard deviation of Ll-RSRP used for training the model.

[0082] At step 5), the UE-side model 410 may determine data categorization based on the received assistance information. The UE 421 may categorize the type of data by comparing the received assistance information, for example, mean and standard deviation of RSRP for current NW configuration with the UE model parameters, for example, statistical measures like mean and standard deviation of the data used for training themodel.

[0083] Alternatively, or additionally, for example in option B, the gNB 420 may categorize the type of data by comparing the received assistance information, for example, statistical proprieties of the data used for training the model with the mean and standard deviation of the beam measurements reported from one or more UEs for current NW configurations.

[0084] At step 6), the UE-side model 410 may perform data preprocessing based on the received assistance information. The assistance information received from gNB 420, for example, mean and standard deviation associated with a specific NW configuration may be used to normalize / standardize the Ll-RSRP values to have zero mean and unitary standard deviation so that they can be used as the input of the UE-side model 412.

[0085] Now reference is made to FIG. 5 to illustrate an example signaling flow 500 for determining the assistance information with specific NW configuration. FIG. 5 shows some steps for determining the assistance information to be used as assistance information signalled from gNB 530 to another UE. It is noted that transmission beams are indicated as the set of beams to be used for model input.

[0086] The gNB 530 may configure (532) the CSI-RSs to UE 510 of the cell and the gNB 530 may transmit (534) the Tx beams corresponding to the CSI-RSs to UE 510 of the cell. The gNB 530 may configure (531) the CSI-RSs to UE 520 of the cell and the gNB 530 may transmit (533) the Tx beams corresponding to the CSI-RSs to UE 520 of the cell.

[0087] Then, the gNB 530 may receive (511) the Ll-RSRP measurements reporting from UE 510. The gNB 530 may receive (521) the Ll-RSRP measurements reporting from UE 520. The gNB 530 may collect a sufficient number of samples to determine the statistical measures associated with the current NW configuration.

[0088] The gNB 530 may determine (535) the assistance information corresponding to the current NW configuration. For instance, the gNB 530 may determine the scalar parameters, such as the mean and standard deviation of the Ll-RSRP measurements.

[0089] In some example embodiments, the mean and standard deviation of the Ll-RSRP measurements determined per NW configuration can be expressed as:where RSRP(m)kis the RSRP for each Set B beam for n = 0, ... ,N and NW configuration k.

[0090] In some example embodiments, the mean and standard deviation of the Ll-RSRP measurements determined per Tx beam, i.e., for each model input feature can be expressedare the mean and standard deviation for each Set B beam for n = 0, ... ,N and gNB 530 configuration k and can be expressed as«: = 1 / M S"=i RSRP(n (3)

[0091] In a similar way, the UE and / or UE server may determine assistance information based on the Ll-RSRP measurements collected at the UE / UE server and used to train the model deployed at the UE.

[0092] The example architecture is described above. Now reference is made to FIG. 6 and FIG. 7 to illustrate example signaling flows for option A and option B, respectively.Example flow with assistance information from NW to UE

[0093] FIG. 6 shows the flow chart 600 depicting the transmission of assistance information from the gNB 602 to the UE 601 for supporting the UE data collection. The UE 601 is an example of the first apparatus 110 and the gNB 602 is an example of the second apparatus 603.

[0094] The UE 601 may transmit (605) a message to request data collection configuration. After receiving the request, the gNB 602 may change the CSI-RSs configuration to allow the UE 601 to measure the Set A beams and Set B beams (in the case that the UE model supports Set B is different from Set A beam prediction) for collecting new beam measurements for updating / fine-tuning / switching the UE-side model.

[0095] The UE 601 may transmit (610) a message indicating the need for receiving assistance information from the gNB 602 to the UE 601.

[0096] The gNB 602 may transmit (615) the CSI-RSs corresponding to the beams in Set A (in the case that the UE model supporting Set B is a subset of Set A beam prediction) to the UE 601. Alternatively, the gNB 602 may transmit (615) the CSI-RS corresponding to the beams in Set A and Set B beams (in the case that the UE model supports Set B is different from Set A beam prediction). In this way, measurement results, which in this example are RSRPs, may be obtained.

[0097] The gNB 602 may transmit (620) the assistance information to support the UE 601 data collection. For instance, the assistance information may include the mean and standard deviation of the Ll-RSRP measurements associated with the current NW configurations used by the gNB 602 for RS transmission. The mean and standard deviation can be determined per NW configuration or per Tx beam, as described above. Alternatively, or in addition, the assistance information may include an identifier of the current NW configurations used by the gNB 602 for RS transmission.

[0098] Based on the received assistance information, the UE 601 may perform (625) data categorization and determines if the data can be used for updating / fine-tuning the UE-side model. It is to be noted that the term “data” used herein means the measurement results mentioned above.

[0099] In some example embodiments, the UE 601 may categorize the data by verifying if the standard scaler parameters, i.e., mean and standard deviation of RSRP associated with the UE model, may be the same of the assistance information, for example, statistical proprieties of the data (mean and standard deviation of RSRP for current NW configuration) received as assistant information from the gNB 602.

[0100] Alternatively, or in addition, in some example embodiments, the received assistance information may include an identifier of the current NW configuration of the gNB 602. The UE 601 may categorize the data by comparing an identifier of an NW configuration associated with the model with the identifier of the current NW configuration.

[0101] The UE 601 may proceed (630) with updating / fine-tuning / switching the UE-side model, either on its own or in a remote UE server. For example, in case of a positive outcome of the data categorization step, the beam measurements collected at the UE 601 may be used for updating / fine-tuning the UE-side model. Otherwise, the beam measurements collected may not be used for updating / fine-tuning the UE-side model andmay be labeled to belong to a different NW configuration. The UE-side model may be switched to another model to be used for current NW configuration.

[0102] In some example embodiments, the UE 601 may use (635) the assistance information received from the gNB 602 for pre-processing the measurement results before using them as input to the UE-side model. In one example, the assistance information received from the gNB 602, for example, the mean and standard deviation associated with the gNB 602 configuration k may be used to normalize or standardize the LI -RSRP values as follows:where RSRP represents the standardized / normalized value of the RSRP. After preprocessing, the range of LI -RSRP values is normalized / standardized to have mean p = 0 and standard deviation o = I and can be used as the input to the UE-side model.Example flow with assistance information from UE to NW

[0103] As mentioned above, in some example embodiments, the assistance information may be transmitted from the UE to the gNB. Reference is now made to FIG. 7. As shown, the assistance information is transmitted from UE 701 to gNB 702 for supporting the UE data collection.

[0104] The UE 701 may transmit (705) a message requesting for data collection configuration (similar as the act 605 as shown in FIG. 6). The gNB 702 may transmit (710) a message indicating the need for transmitting assistance information from UE 701 to gNB 702.

[0105] The gNB 702 may transmit (715) the CSI-RSs to the UE 71 (similar to the act 615 as shown in FIG. 6). As a result, measurement results are obtained by the UE 701. e UE 701 may transmit (720) the measurements results to the gNB 702.

[0106] The UE 701 may transmit (725) the assistance information to gNB 702. For instance, the assistance information may include the mean and standard deviation of the LI -RSRP measurements associated with the current UE model deployed at the UE 701.

[0107] The gNB 702 may perform (730) data categorization based on the received assistance information and other assistance information determines on its own. The gNB 702 may determine if the data (referring to the measurement results received at act 720)can be used for updating / fine-tuning the UE-side model.

[0108] For example, the gNB 702 may categorize the data by verifying if the assistance information, for example mean and standard deviation of RSRPs used to train the UE model, have the same statistical proprieties of the data for current NW configuration measured from the data reported from the UE 701. For another example, the assistance information from the UE 701 may include an identifier of an NW configuration associated with the model used at the UE 701. The gNB 702 may categorize the data by comparing the identifier of the NW configuration associated with the model with an identifier of the current NW configuration of the gNB 702.

[0109] The gNB 702 may indicate (735) to the UE 701 whether the data received at act 725 can be used for updating / fine-tuning the UE-side model. In case of a positive “data collection indication”, the data, which refers to the measurement results obtained at 715, may be used for updating / fine-tuning the UE-side model.

[0110] In case of a negative “data collection indication”, the beam measurements collected may not be used for updating / fine-tuning the UE-side model and may be labelled to belong to a different NW configuration. In case of a negative “data collection indication” the UE-side model may be switched to another model to be used for current NW configuration.

[0111] In some example embodiment, the “data collection indication” may also provide a soft indication about which measurements to use for updating / fine-tuning the UE-side model and which measurements not to use.

[0112] The UE 701 may proceed (740) with updating / fine-tuning / switching the UE-side model, similarly to what is described with respect the act 730 as shown in FIG. 7.

[0113] For assisted data collection for model update at UE server, in some example, the use of assistance information for UE 821 and UE 831 data collection when the UE 821 and UE 831 sends the beam measurements to an external UE server 810, which performs the necessary steps for updating the model.Assisted data collection for model update at UE server

[0114] As mentioned above, in some example embodiment, update of the UE side model may be performed at a UE server. Reference is now made to FIG. 8. As shown, two gNBs(e.g., gNB 820 and gNB 830) sends (801) assistance information to the UEs (e.g., UE 821 and UE 831) served in their respective cells to support data collection. The two gNBs may employ different configurations of the antenna pattern. Therefore, the assistance information sent to the UEs have mean pi and standard deviation oi for the first gNB 820 and mean 2 and standard deviation 02 for the second gNB 830.

[0115] The UEs send (802, 806) the beam measurement results together with the assistance information received from respective gNB to an external UE server 810.

[0116] The UE server 810 may categorize the data depending on the assistance information received. For example, the UE server 810 may categorize the beam measurements collected with the gNB 820 as a first type of data, represented by data 811 with dashed line. The UE server 810 may categorize the beam measurement results collected with the gNB 830 as a second type of data represented by data 812 with dotted line.

[0117] The UE server 810 may use (804) the assistance information in the preprocessing step, to scale / normalize the RSRPs to have zero mean and unit variance. As the data used are beam measurements collected with the gNB 820, the scaler parameters are the mean pi and standard deviation 01 for the gNB 820.

[0118] The model for the cell served by the gNB 820 is updated. The UE server 810 may use (803) as training dataset 813 only the data of the first type. The model specific for gNB 820 is then updated with this training dataset.

[0119] After the model update, the UE server 810 may transfer / download (805) the updated model to the UEs 821 served by the gNB 820.

[0120] The operation may be repeated for updating the model to be used by the UE 831 served by the gNB 830 by using as training dataset the beam measurements of the second type.Example Signaling

[0121] Some example embodiments are described with respect to the use case of beam prediction. However, it is to be understood that the concept described above may be applicable to any suitable AI / ML use cases, for example CSI prediction, positioning, etc.

[0122] Reference is made to FIG. 9, which illustrates an example signaling flow 900for data collection for AI / ML model according to some example embodiments of the present disclosure. For the purposes of discussion, the signaling flow 900 will be discussed with reference to FIG. 1, for example, by using the first apparatus 110 and the second apparatus 120.

[0123] As shown, in some embodiments, before data collection, the first apparatus 110 may transmit (905) to the second apparatus 120, a request to configure the first apparatus 110 for measurement result collection. For example, the UE may request the gNB to configure the UE for data collection to update a UE-side model. Accordingly, the first apparatus 110 may receive (910) from the second apparatus 120, a configuration of the set of reference signals. For example, in the use case of beam prediction or CSI prediction, CSI RSs may be configured to the first apparatus 110.

[0124] In some example embodiments, assistance information may be transmitted from the second apparatus 120 to the first apparatus 110, which means option A is employed. In this case, the process 901 is performed.

[0125] In the process 901, the first apparatus 110 may transmit (915) to the second apparatus 120, a first message indicating a need for assistance to categorize measurement results. For example, the UE may request assistance information from the gNB.

[0126] The second apparatus 120 may transmit (920) to a first apparatus 110, the set of reference signals for obtaining a first set of measurement results to be used to a model. For example, CSI RSs may be transmitted to the first apparatus 110.

[0127] Accrordingly, the first apparatus 110 may obtain (925), by measuring a set of reference signals from the second apparatus 120, a first set of measurement results to be used to the model. For example, the use case of beam prediction, the first set of measurement results may be Ll-RSRPs, as described above.

[0128] The second apparatus 120 may transmit (930) first assistance information from the first apparatus 110. The first assistance information indicates a first characteristic of a second set of measurement results associated with the set of reference signals. To determine the first assistance information, the second apparatus 120 may collect the second set of measurement results from a plurality of apparatuses. The second set of measurement results are obtained with the same configuration as the set of reference signals. Then, the second apparatus 120 may determine the first assistance informationbased on the second set of measurement results. For example, the first assistance information may be determined as described with reference to FIG. 5.

[0129] In some example embodiments, the first characteristic comprises at least one of: a mean of the second set of measurement results, a standard deviation of the second set of measurement results, or an identifier of a configuration of the second apparatus for obtaining the second set of measurement results.

[0130] In some example embodiments, the first characteristic may be determined and thus indicated per a configuration of the second apparatus to transmit the set of reference signals, for example, per NW configuration as described above. In some example embodiments, the first characteristic may be determined and thus indicated per a transmitting beam corresponding to a respective reference signal of the set of reference signals, for example, per Tx beam as described above.

[0131] After receiving the first assistance information, the first apparatus 110 may compare (935) the first characteristic with a second characteristic of a third set of measurement results used for training the model. For example, the second characteristic may be determined by the first apparatus 110 if the first apparatus 110 trains the model. Alternatively, the second characteristic may be indicated by the third apparatus 130 to the first apparatus 110.

[0132] In some example embodiments, assistance information may be transmitted from the first apparatus 110 to the second apparatus 120, which means option B is employed. In this case, the process 902 is performed.

[0133] In the process 902, the second apparatus 120 may transmit (940) to the first apparatus 110, a second message indicating a need for assistance to categorize the first set of measurement results.

[0134] The second apparatus 120 may transmit (920) to a first apparatus 110, the set of reference signals for obtaining a first set of measurement results to be used to a model. For example, CSI RSs may be transmitted to the first apparatus 110.

[0135] Accrordingly, the first apparatus 110 may obtain (925), by measuring a set of reference signals from the second apparatus 120, a first set of measurement results to be used to the model. For example, the use case of beam prediction, the first set of measurement results may be Ll-RSRPs, as described above.

[0136] Then, the first apparatus 110 may transmit (945) the first set of measurement results and 950) to the second apparatus 120, and transmit (945) second assistance information indicating the second characteristic of the third set of measurement results used for training the model.

[0137] The second apparatus 120 may determine (955) the indication of the usability of the first set of measurement results as input to the model by comparing the first characteristic with the second characteristic. The indication may indicate at least one of: whether the first set of measurement results are usable for the model, one or more measurement results usable for the model, or one or more measurement results unusable for the model.

[0138] In some example embodiments, the second apparatus 120 may determine whether the first set of measurement results have a same data type as the third set of measurement results based on the comparing. If the first set of measurement results have the same data type as the third set of measurement results, the second apparatus 120 may generate the indication as that the first set of measurement results are usable for the model. If the first set of measurement results have a different data type from the third set of measurement results, the second apparatus 120 may generate the indication as that the first set of measurement results are unusable for the model.

[0139] The first apparatus 110 may receive (960), from the second apparatus 120, the indication of the usability of the first set of measurement results as input to the model.

[0140] Based on the result of data categorization at 935 or the indication from the second apparatus 120, the first apparatus 110 may perform (965) an action associated with the first set of measurement results on the model based on the at least one of the first assistance information or the indication.

[0141] In some example embodiments, if the first set of measurement results have the same data type as the third set of measurement results or the indication indicates that the first set of measurement results are usable for the model, the first apparatus 110 may update the model based on the first set of measurement results, for example, retraining, or fine-tuning the model. If the first set of measurement results have a different data type from the third set of measurement results or the indication indicates that the first set of measurement results are unusable for the model, the first apparatus 110 may switch, based on the first assistance information, from the model to another model for obtaining aprediction by using the first set of measurement results as input.

[0142] If the first apparatus 110 updates the model, the first apparatus 110 may pre- process (970) the first set of measurement results based on the first assistance information before using the first set of measurement results as inputs to the model. For example, the first set of measurement results may be normalized or standardized based on the mean and standard deviation of the second set of measurement results.

[0143] Alternatively, or in addition, the first apparatus 110 may transmit (975) the first set of measurement results and the first assistance information to the third apparatus 130 for updating the model. The third apparatus 130 may determine (980) a data type of the first set of measurement results based on the first assistance information. The third apparatus 130 may update (985) a model corresponding to the data type of the first set of measurement results based on the first set of measurement results.

[0144] As mentioned above, in some example embodiments, the second assistance information may be generated by the third apparatus 130. As shown, the third apparatus 130 may determine (927) the second characteristic of the third set of measurement results used for training the model. The third apparatus 130 may transmit (947) to the first apparatus 110, the second assistance information indicating the second characteristic of the third set of measurement results.Example Methods

[0145] FIG. 10 shows a flowchart of an example method 1000 implemented at a first apparatus in accordance with some example embodiments of the present disclosure. For the purpose of discussion, the method 1000 will be described from the perspective of the first apparatus 110 in FIG. 1.

[0146] At block 1010, the first apparatus obtains, by measuring a set of reference signals from a second apparatus, a first set of measurement results to be used to a model.

[0147] At block 1020, the first apparatus receives, from the second apparatus, at least one of: first assistance information indicating a first characteristic of a second set of measurement results associated with the set of reference signals, or an indication of a usability of the first set of measurement results as input to the model.

[0148] At block 1030, the first apparatus performs an action associated with the first set of measurement results on the model based on the at least one of the first assistanceinformation or the indication.

[0149] In some example embodiments, the first apparatus may transmit, to the second apparatus, a first message indicating a need for assistance to categorize the first set of measurement results; and receive the first assistance information from the second apparatus.

[0150] In some example embodiments, the first apparatus may compare the first characteristic with a second characteristic of a third set of measurement results used for training the model; determine whether the first set of measurement results have a same data type as the third set of measurement results based on the comparing; and in accordance with a determination that the first set of measurement results have the same data type as the third set of measurement results, update the model based on the first set of measurement results.

[0151] In some example embodiments, the first apparatus may pre-process the first set of measurement results based on the first assistance information before using the first set of measurement results as inputs to the model.

[0152] In some example embodiments, the first apparatus may transmit the first set of measurement results and the first assistance information to a third apparatus for updating the model.

[0153] In some example embodiments, in accordance with a determination that the first set of measurement results have a different data type from the third set of measurement results, the first apparatus may switch, based on the first assistance information, from the model to another model for obtaining a prediction by using the first set of measurement results as input.

[0154] In some example embodiments, the first apparatus may transmit, to the second apparatus, the first set of measurement results and second assistance information indicating a second characteristic of a third set of measurement results used for training the model; and receive, from the second apparatus, the indication of the usability of the first set of measurement results as input to the model.

[0155] In some example embodiments, the first apparatus may receive, from the second apparatus, a second message indicating a need for assistance to categorize the first set of measurement results.

[0156] In some example embodiments, the indication may indicate at least one of: whether the first set of measurement results are usable for the model, one or more measurement results usable for the model, or one or more measurement results unusable for the model.

[0157] In some example embodiments, the first apparatus may switch, based on the indication, from the model to another model for obtaining a prediction by using the first set of measurement results as input.

[0158] In some example embodiments, the first assistance information may be received via at least one of: a radio resource control signaling, or a MAC CE.

[0159] In some example embodiments, the set of reference signals correspond to different transmitting beams, the first set of measurement results comprise reference signal received powers, and the model is used for beam prediction.

[0160] In some example embodiments, the first apparatus may transmit, to the second apparatus, a request to configure the first apparatus for measurement result collection; and receiving, from the second apparatus, a configuration of the set of reference signals.

[0161] In some example embodiments, the first characteristic comprises at least one of: a mean of the second set of measurement results, a standard deviation of the second set of measurement results, or an identifier of a configuration of the second apparatus for obtaining the second set of measurement results.

[0162] In some example embodiments, the first characteristic is indicated per at least one of: a configuration of the second apparatus to transmit the set of reference signals, or a transmitting beam corresponding to a respective reference signal of the set of reference signals.

[0163] In some example embodiments, the first apparatus may receive, from a third apparatus, second assistance information indicating the second characteristic of the third set of measurement results used for training the model.

[0164] In some example embodiments, the first apparatus comprises a terminal device, and the second apparatus comprises a network device.

[0165] FIG. 11 shows a flowchart of an example method 1100 implemented at a second apparatus in accordance with some example embodiments of the present disclosure. Forthe purpose of discussion, the method 1100 will be described from the perspective of the second apparatus 120 in FIG. 1.

[0166] At block 1110, the second apparatus transmits, to a first apparatus, a set of reference signals for obtaining a first set of measurement results to be used to a model.

[0167] At block 1120, the second apparatus transmits, to the first apparatus, at least one of: first assistance information indicating a first characteristic of a second set of measurement results associated with the set of reference signals, or an indication of a usability of the first set of measurement results as input to the model.

[0168] In some example embodiments, the second apparatus transmits receiving, from the first apparatus, a first message indicating a need for assistance to categorize the first set of measurement results; and transmitting the first assistance information to the first apparatus.

[0169] In some example embodiments, the second apparatus may collect the second set of measurement results from a plurality of apparatuses, wherein the second set of measurement results are obtained with the same configuration as the set of reference signals; and determine the first assistance information based on the second set of measurement results.

[0170] In some example embodiments, the first characteristic is determined per at least one of: a configuration of the second apparatus to transmit the set of reference signals, or a transmitting beam corresponding to a respective reference signal of the set of reference signals.

[0171] In some example embodiments, the second apparatus may receive, from the first apparatus, the first set of measurement results and second assistance information indicating a second characteristic of a third set of measurement results used for training the model; determine the indication of the usability of the first set of measurement results as input to the model by comparing the first characteristic with the second characteristic; and transmit the indication to the first apparatus.

[0172] In some example embodiments, the second apparatus may determine whether the first set of measurement results have a same data type as the third set of measurement results based on the comparing; in accordance with a determination that the first set of measurement results have the same data type as the third set of measurement results,generate the indication as that the first set of measurement results are usable for the model; and in accordance with a determination that the first set of measurement results have a different data type from the third set of measurement results, generate the indication as that the first set of measurement results are unusable for the model.

[0173] In some example embodiments, the second apparatus transmits, to the first apparatus, a second message indicating a need for assistance to categorize the first set of measurement results.

[0174] In some example embodiments, the indication may indicate at least one of: whether the first set of measurement results are usable for the model, one or more measurement results usable for the model, or one or more measurement results unusable for the model.

[0175] In some example embodiments, the first assistance information is transmitted via at least one of: a radio resource control signaling, or a MAC CE.

[0176] In some example embodiments, the set of reference signals correspond to different transmitting beams, the first set of measurement results comprise reference signal received powers, and the model is used for beam prediction.

[0177] In some example embodiments, the second apparatus may receive, from the first apparatus, a request to configure the first apparatus for measurement result collection; and transmit, to the first apparatus, a configuration of the set of reference signals.

[0178] In some example embodiments, the first characteristic comprises at least one of: a mean of the second set of measurement results, a standard deviation of the second set of measurement results, or an identifier of a configuration of the second apparatus for obtaining the second set of measurement results.

[0179] In some example embodiments, the first apparatus comprises a terminal device, and the second apparatus comprises a network device.

[0180] FIG. 12 shows a flowchart of an example method 1200 implemented at a third apparatus in accordance with some example embodiments of the present disclosure. For the purpose of discussion, the method 1200 will be described from the perspective of the third apparatus 130 in FIG. 1.

[0181] At block 1210, the third apparatus receives, from a first apparatus, a first set ofmeasurement results on a set of reference signals from a second apparatus and first assistance information indicating a first characteristic of a second set of measurement results associated with the set of reference signals.

[0182] At block 1220, the third apparatus determines a data type of the first set of measurement results based on the first assistance information.

[0183] At block 1230, the third apparatus updates a model corresponding to the data type of the first set of measurement results based on the first set of measurement results.

[0184] In some example embodiments, the third apparatus may pre-process the first set of measurement results based on the first assistance information before using the first set of measurement results as inputs to the model.

[0185] In some example embodiments, the first characteristic comprises at least one of: a mean of the second set of measurement results, a standard deviation of the second set of measurement results, or an identifier of a configuration of the second apparatus for obtaining the second set of measurement results.

[0186] In some example embodiments, the third apparatus may determine a second characteristic of a third set of measurement results used for training the model; and transmit, to the first apparatus, second assistance information indicating the second characteristic of the third set of measurement results.Example Apparatus, Device and Medium

[0187] In some example embodiments, a first apparatus capable of performing any of the method 1000 (for example, the first apparatus 110 in FIG. 1 may comprise means for performing the respective operations of the method 1000. 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.

[0188] In some example embodiments, the first apparatus comprises means for obtaining, by measuring a set of reference signals from a second apparatus, a first set of measurement results to be used to a model; means for receiving, from the second apparatus, at least one of: first assistance information indicating a first characteristic of a second set of measurement results associated with the set of reference signals, or an indication of a usability of the first set of measurement results as input to the model; and means forperforming an action associated with the first set of measurement results on the model based on the at least one of the first assistance information or the indication.

[0189] In some example embodiments, the first apparatus comprises: means for transmitting, to the second apparatus, a first message indicating a need for assistance to categorize the first set of measurement results; and means for receiving the first assistance information from the second apparatus.

[0190] In some example embodiments, the first apparatus comprises: means for comparing the first characteristic with a second characteristic of a third set of measurement results used for training the model; means for determining whether the first set of measurement results have a same data type as the third set of measurement results based on the comparing; and means for in accordance with a determination that the first set of measurement results have the same data type as the third set of measurement results, updating the model based on the first set of measurement results.

[0191] In some example embodiments, the first apparatus comprises: means for preprocessing the first set of measurement results based on the first assistance information before using the first set of measurement results as inputs to the model.

[0192] In some example embodiments, the first apparatus comprises: means for transmitting the first set of measurement results and the first assistance information to a third apparatus for updating the model.

[0193] In some example embodiments, the first apparatus comprises: means for in accordance with a determination that the first set of measurement results have a different data type from the third set of measurement results, switching, based on the first assistance information, from the model to another model for obtaining a prediction by using the first set of measurement results as input.

[0194] In some example embodiments, the first apparatus comprises: means for transmitting, to the second apparatus, the first set of measurement results and second assistance information indicating a second characteristic of a third set of measurement results used for training the model; and means for receiving, from the second apparatus, the indication of the usability of the first set of measurement results as input to the model.

[0195] In some example embodiments, the first apparatus comprises: means for receiving, from the second apparatus, a second message indicating a need for assistanceto categorize the first set of measurement results.

[0196] In some example embodiments, the indication may indicate at least one of: whether the first set of measurement results are usable for the model, one or more measurement results usable for the model, or one or more measurement results unusable for the model.

[0197] In some example embodiments, the first apparatus comprises: means for switching, based on the indication, from the model to another model for obtaining a prediction by using the first set of measurement results as input.

[0198] In some example embodiments, the first assistance information is received via at least one of: a radio resource control signaling, or a MAC CE.

[0199] In some example embodiments, the set of reference signals correspond to different transmitting beams, the first set of measurement results comprise reference signal received powers, and the model is used for beam prediction.

[0200] In some example embodiments, the first apparatus comprises: means for transmitting, to the second apparatus, a request to configure the first apparatus for measurement result collection; and means for receiving, from the second apparatus, a configuration of the set of reference signals.

[0201] In some example embodiments, the first characteristic comprises at least one of: a mean of the second set of measurement results, a standard deviation of the second set of measurement results, or an identifier of a configuration of the second apparatus for obtaining the second set of measurement results.

[0202] In some example embodiments, the first characteristic is indicated per at least one of: a configuration of the second apparatus to transmit the set of reference signals, or a transmitting beam corresponding to a respective reference signal of the set of reference signals.

[0203] In some example embodiments, the first apparatus comprises: means for receiving, from a third apparatus, second assistance information indicating the second characteristic of the third set of measurement results used for training the model.

[0204] In some example embodiments, the first apparatus comprises a terminal device, and the second apparatus comprises a network device.

[0205] In some example embodiments, the first apparatus further comprises means for performing other operations in some example embodiments of the method 1000 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.

[0206] In some example embodiments, a second apparatus capable of performing any of the method 1100 (for example, the second apparatus 120 in FIG. 1) may comprise means for performing the respective operations of the method 1100. 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.

[0207] In some example embodiments, the second apparatus comprises means for transmitting, to a first apparatus, a set of reference signals for obtaining a first set of measurement results to be used to a model; and means for transmitting, to the first apparatus, at least one of: first assistance information indicating a first characteristic of a second set of measurement results associated with the set of reference signals, or an indication of a usability of the first set of measurement results as input to the model.

[0208] In some example embodiments, the second apparatus comprises: means for receiving, from the first apparatus, a first message indicating a need for assistance to categorize the first set of measurement results; and means for transmitting the first assistance information to the first apparatus.

[0209] In some example embodiments, the second apparatus comprises: means for collecting the second set of measurement results from a plurality of apparatuses, wherein the second set of measurement results are obtained with the same configuration as the set of reference signals; and means for determining the first assistance information based on the second set of measurement results.

[0210] In some example embodiments, the first characteristic is determined per at least one of: a configuration of the second apparatus to transmit the set of reference signals, or a transmitting beam corresponding to a respective reference signal of the set of reference signals.

[0211] In some example embodiments, the second apparatus comprises: means forreceiving, from the first apparatus, the first set of measurement results and second assistance information indicating a second characteristic of a third set of measurement results used for training the model; means for determining the indication of the usability of the first set of measurement results as input to the model by comparing the first characteristic with the second characteristic; and means for transmitting the indication to the first apparatus.

[0212] In some example embodiments, the second apparatus comprises: means for determining whether the first set of measurement results have a same data type as the third set of measurement results based on the comparing; means for in accordance with a determination that the first set of measurement results have the same data type as the third set of measurement results, generating the indication as that the first set of measurement results are usable for the model; and means for in accordance with a determination that the first set of measurement results have a different data type from the third set of measurement results, generating the indication as that the first set of measurement results are unusable for the model.

[0213] In some example embodiments, the second apparatus comprises: means for transmitting, to the first apparatus, a second message indicating a need for assistance to categorize the first set of measurement results.

[0214] In some example embodiments, the indication indicates at least one of: whether the first set of measurement results are usable for the model, one or more measurement results usable for the model, or one or more measurement results unusable for the model.

[0215] In some example embodiments, the first assistance information is transmitted via at least one of: a radio resource control signaling, or an MAC CE.

[0216] In some example embodiments, the set of reference signals correspond to different transmitting beams, the first set of measurement results comprise reference signal received powers, and the model is used for beam prediction.

[0217] In some example embodiments, the second apparatus comprises: means for receiving, from the first apparatus, a request to configure the first apparatus for measurement result collection; and means for transmitting, to the first apparatus, a configuration of the set of reference signals.

[0218] In some example embodiments, the first characteristic comprises at least one of:a mean of the second set of measurement results, a standard deviation of the second set of measurement results, or an identifier of a configuration of the second apparatus for obtaining the second set of measurement results.

[0219] In some example embodiments, the first apparatus comprises a terminal device, and the second apparatus comprises a network device.

[0220] In some example embodiments, the second apparatus further comprises means for performing other operations in some example embodiments of the method 1100 or the second 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.

[0221] In some example embodiments, a third apparatus capable of performing any of the method 1200 (for example, the third apparatus 130 in FIG. 1) may comprise means for performing the respective operations of the method 1200. The means may be implemented in any suitable form. For example, the means may be implemented in a circuitry or software module. The third apparatus may be implemented as or included in the third apparatus 130 in FIG. 1.

[0222] In some example embodiments, the third apparatus comprises means for receiving, from a first apparatus, a first set of measurement results on a set of reference signals from a second apparatus and first assistance information indicating a first characteristic of a second set of measurement results associated with the set of reference signals; means for determining a data type of the first set of measurement results based on the first assistance information; and means for updating a model corresponding to the data type of the first set of measurement results based on the first set of measurement results.

[0223] In some example embodiments, the third apparatus comprises: means for preprocessing the first set of measurement results based on the first assistance information before using the first set of measurement results as inputs to the model.

[0224] In some example embodiments, the first characteristic comprises at least one of: a mean of the second set of measurement results, a standard deviation of the second set of measurement results, or an identifier of a configuration of the second apparatus for obtaining the second set of measurement results.

[0225] In some example embodiments, the third apparatus comprises: means for determining a second characteristic of a third set of measurement results used for training the model; and means for transmitting, to the first apparatus, second assistance information indicating the second characteristic of the third set of measurement results.

[0226] In some example embodiments, the third apparatus further comprises means for performing other operations in some example embodiments of the method 1200 or the third apparatus 130. 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 third apparatus.

[0227] FIG. 13 is a simplified block diagram of a device 1300 that is suitable for implementing example embodiments of the present disclosure. The device 1300 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 1300 includes one or more processors 1310, one or more memories 1320 coupled to the processor 1310, and one or more communication modules 1340 coupled to the processor 1310.

[0228] The communication module 1340 is for bidirectional communications. The communication module 1340 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 1340 may include at least one antenna.

[0229] The processor 1310 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 1300 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.

[0230] The memory 1320 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) 1324, 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) 1322 and other volatile memories that will not last in the power-down duration.

[0231] A computer program 1330 includes computer executable instructions that are executed by the associated processor 1310. The instructions of the program 1330 may include instructions for performing operations / acts of some example embodiments of the present disclosure. The program 1330 may be stored in the memory, e.g., the ROM 1324. The processor 1310 may perform any suitable actions and processing by loading the program 1330 into the RAM 1322.

[0232] The example embodiments of the present disclosure may be implemented by means of the program 1330 so that the device 1300 may perform any process of the disclosure as discussed with reference to FIG. 5 to FIG. 9. The example embodiments of the present disclosure may also be implemented by hardware or by a combination of software and hardware.

[0233] In some example embodiments, the program 1330 may be tangibly contained in a computer readable medium which may be included in the device 1300 (such as in the memory 1320) or other storage devices that are accessible by the device 1300. The device 1300 may load the program 1330 from the computer readable medium to the RAM 1322 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).

[0234] FIG. 14 shows an example of the computer readable medium 1400 which may be in form of CD, DVD or other optical storage disk. The computer readable medium 1400 has the program 1330 stored thereon.

[0235] 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 pictorialrepresentations, 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.

[0236] 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.

[0237] 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.

[0238] 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.

[0239] 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 specificexamples 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.

[0240] 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. Eikewise, 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.

[0241] 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

WHAT IS CLAIMED IS:

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: obtain, by measuring a set of reference signals from a second apparatus, a first set of measurement results to be used to a model; receive, from the second apparatus, at least one of: first assistance information indicating a first characteristic of a second set of measurement results associated with the set of reference signals, or an indication of a usability of the first set of measurement results as input to the model; and perform an action associated with the first set of measurement results on the model based on the at least one of the first assistance information or the indication.

2. The first apparatus of claim 1, wherein the instructions, when executed by the at least one processor, cause the first apparatus to: transmit, to the second apparatus, a first message indicating a need for assistance to categorize the first set of measurement results; and receive the first assistance information from the second apparatus.

3. The first apparatus of claim 1 or 2, wherein the instructions, when executed by the at least one processor, cause the first apparatus to: comparing the first characteristic with a second characteristic of a third set of measurement results used for training the model; determine whether the first set of measurement results have a same data type as the third set of measurement results based on the comparing; and in accordance with a determination that the first set of measurement results have the same data type as the third set of measurement results, update the model based on the first set of measurement results.

4. The first apparatus of claim 3, wherein the instructions, when executed by the at least one processor, cause the first apparatus to: pre-process the first set of measurement results based on the first assistance information before using the first set of measurement results as inputs to the model.

5. The first apparatus of claim 3 or 4, wherein the instructions, when executed by the at least one processor, cause the first apparatus to: transmit the first set of measurement results and the first assistance information to a third apparatus for updating the model.

6. The first apparatus of any of claims 3-5, wherein the instructions, when executed by the at least one processor, cause the first apparatus to: in accordance with a determination that the first set of measurement results have a different data type from the third set of measurement results, switch, based on the first assistance information, from the model to another model for obtaining a prediction by using the first set of measurement results as input.

7. The first apparatus of any of claims 1-6, wherein the instructions, when executed by the at least one processor, cause the first apparatus to: transmit, to the second apparatus, the first set of measurement results and second assistance information indicating a second characteristic of a third set of measurement results used for training the model; and receive, from the second apparatus, the indication of the usability of the first set of measurement results as input to the model.

8. The first apparatus of claim 7, wherein the instructions, when executed by the at least one processor, cause the first apparatus to: receive, from the second apparatus, a second message indicating a need for assistance to categorize the first set of measurement results.

9. The first apparatus of any of claims 1-8, wherein the indication indicates at least one of: whether the first set of measurement results are usable for the model, one or more measurement results usable for the model, orone or more measurement results unusable for the model.

10. The first apparatus of any of claims 1-9, wherein the instructions, when executed by the at least one processor, cause the first apparatus to: switch, based on the indication, from the model to another model for obtaining a prediction by using the first set of measurement results as input.

11. The first apparatus of any of claims 1-10, wherein the first assistance information is received via at least one of: a radio resource control signaling, or a medium access control control element.

12. The first apparatus of any of claims 1-11, wherein the set of reference signals correspond to different transmitting beams, the first set of measurement results comprise reference signal received powers, and the model is used for beam prediction.

13. The first apparatus of any of claims 1-12, wherein the instructions, when executed by the at least one processor, cause the first apparatus to: transmit, to the second apparatus, a request to configure the first apparatus for measurement result collection; and receive, from the second apparatus, a configuration of the set of reference signals.

14. The first apparatus of any of claims 1-13, wherein the first characteristic comprises at least one of: a mean of the second set of measurement results, a standard deviation of the second set of measurement results, or an identifier of a configuration of the second apparatus for obtaining the second set of measurement results.

15. The first apparatus of any of claims 1-14, wherein the first characteristic is indicated per at least one of: a configuration of the second apparatus to transmit the set of reference signals, or a transmitting beam corresponding to a respective reference signal of the set of reference signals.

16. The first apparatus of any of claims 3-5 or 7-8, wherein the instructions, when executed by the at least one processor, cause the first apparatus to: receive, from a third apparatus, second assistance information indicating the second characteristic of the third set of measurement results used for training the model.

17. The first apparatus of any of claims 1-6, wherein the first apparatus comprises a terminal device, and the second apparatus comprises a network device.

18. 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, a set of reference signals for obtaining a first set of measurement results to be used to a model; and transmit, to the first apparatus, at least one of: first assistance information indicating a first characteristic of a second set of measurement results associated with the set of reference signals, or an indication of a usability of the first set of measurement results as input to the model.

19. The second apparatus of claim 18, wherein the instructions, when executed by the at least one processor, cause the second apparatus to: receive, from the first apparatus, a first message indicating a need for assistance to categorize the first set of measurement results; and transmit the first assistance information to the first apparatus.

20. The second apparatus of claim 19, wherein the instructions, when executed by the at least one processor, cause the second apparatus to: collect the second set of measurement results from a plurality of apparatuses, wherein the second set of measurement results are obtained with the same configuration as the set of reference signals; and determine the first assistance information based on the second set of measurement results.

21. The second apparatus of claim 20, wherein the first characteristic is determined per at least one of: a configuration of the second apparatus to transmit the set of reference signals, or a transmitting beam corresponding to a respective reference signal of the set of reference signals.

22. The second apparatus of any of claims 18-21, wherein the instructions, when executed by the at least one processor, cause the second apparatus to: receive, from the first apparatus, the first set of measurement results and second assistance information indicating a second characteristic of a third set of measurement results used for training the model; determine the indication of the usability of the first set of measurement results as input to the model by comparing the first characteristic with the second characteristic; and transmit the indication to the first apparatus.

23. The second apparatus of claim 22, wherein the instructions, when executed by the at least one processor, cause the second apparatus to: determine whether the first set of measurement results have a same data type as the third set of measurement results based on the comparing; in accordance with a determination that the first set of measurement results have the same data type as the third set of measurement results, generate the indication as that the first set of measurement results are usable for the model; and in accordance with a determination that the first set of measurement results have a different data type from the third set of measurement results, generate the indication as that the first set of measurement results are unusable for the model.

24. The second apparatus of claim 22, wherein the instructions, when executed by the at least one processor, cause the second apparatus to: transmit, to the first apparatus, a second message indicating a need for assistance to categorize the first set of measurement results.

25. The second apparatus of any of claims 18-24, wherein the indication indicates at least one of:whether the first set of measurement results are usable for the model, one or more measurement results usable for the model, or one or more measurement results unusable for the model.

26. The second apparatus of any of claims 18-25, wherein the first assistance information is transmitted via at least one of: a radio resource control signaling, or a medium access control control element.

27. The second apparatus of any of claims 18-26, wherein the set of reference signals correspond to different transmitting beams, the first set of measurement results comprise reference signal received powers, and the model is used for beam prediction.

28. The second apparatus of any of claims 18-27, wherein the instructions, when executed by the at least one processor, cause the second apparatus to: receive, from the first apparatus, a request to configure the first apparatus for measurement result collection; and transmit, to the first apparatus, a configuration of the set of reference signals.

29. The second apparatus of any of claims 18-28, wherein the first characteristic comprises at least one of: a mean of the second set of measurement results, a standard deviation of the second set of measurement results, or an identifier of a configuration of the second apparatus for obtaining the second set of measurement results.

30. The second apparatus of any of claims 18-29, wherein the first apparatus comprises a terminal device, and the second apparatus comprises a network device.

31. A third apparatus comprising: at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the third apparatus to:receive, from a first apparatus, a first set of measurement results on a set of reference signals from a second apparatus and first assistance information indicating a first characteristic of a second set of measurement results associated with the set of reference signals; determine a data type of the first set of measurement results based on the first assistance information; and update a model corresponding to the data type of the first set of measurement results based on the first set of measurement results.

32. The third apparatus of claim 31, wherein the instructions, when executed by the at least one processor, cause the third apparatus to: pre-process the first set of measurement results based on the first assistance information before using the first set of measurement results as inputs to the model.

33. The third apparatus of claim 31 or 32, wherein the first characteristic comprises at least one of: a mean of the second set of measurement results, a standard deviation of the second set of measurement results, or an identifier of a configuration of the second apparatus for obtaining the second set of measurement results.

34. The third apparatus of any of claims 31-33, wherein the instructions, when executed by the at least one processor, cause the third apparatus to: determine a second characteristic of a third set of measurement results used for training the model; and transmit, to the first apparatus, second assistance information indicating the second characteristic of the third set of measurement results.

35. A method comprising: obtaining, at a first apparatus by measuring a set of reference signals from a second apparatus, a first set of measurement results to be used to a model; receiving, from the second apparatus, at least one of: first assistance information indicating a first characteristic of a second set of measurement results associated with the set of reference signals, oran indication of a usability of the first set of measurement results as input to the model; and performing an action associated with the first set of measurement results on the model based on the at least one of the first assistance information or the indication.

36. A method comprising: transmitting, at a second apparatus to a first apparatus, a set of reference signals for obtaining a first set of measurement results to be used to a model; and transmitting, to the first apparatus, at least one of: first assistance information indicating a first characteristic of a second set of measurement results associated with the set of reference signals, or an indication of a usability of the first set of measurement results as input to the model.

37. A method comprising: receiving, at a third apparatus from a first apparatus, a first set of measurement results on a set of reference signals from a second apparatus and first assistance information indicating a first characteristic of a second set of measurement results associated with the set of reference signals; determining a data type of the first set of measurement results based on the first assistance information; and updating a model corresponding to the data type of the first set of measurement results based on the first set of measurement results.

38. A first apparatus comprising: means for obtaining (1010), by measuring a set of reference signals from a second apparatus, a first set of measurement results to be used to a model; means for receiving (1020), from the second apparatus, at least one of: first assistance information indicating a first characteristic of a second set of measurement results associated with the set of reference signals, or an indication of a usability of the first set of measurement results as input to the model; and means for performing (1030) an action associated with the first set of measurement results on the model based on the at least one of the first assistance information or the indication.

39. A second apparatus comprising: means for transmitting (1110), to a first apparatus, a set of reference signals for obtaining a first set of measurement results to be used to a model; and means for transmitting (1120), to the first apparatus, at least one of: first assistance information indicating a first characteristic of a second set of measurement results associated with the set of reference signals, or an indication of a usability of the first set of measurement results as input to the model.

40. A third apparatus comprising: means for receiving (1210), from a first apparatus, a first set of measurement results on a set of reference signals from a second apparatus and first assistance information indicating a first characteristic of a second set of measurement results associated with the set of reference signals; means for determining (1220) a data type of the first set of measurement results based on the first assistance information; and means for updating (1230) a model corresponding to the data type of the first set of measurement results based on the first set of measurement results.

41. A computer readable medium comprising instructions stored thereon for causing an apparatus at least to perform the method of any of claims 35-37.

42. A computer program comprising instructions which, when the program is executed by a computer, cause the computer to perform the method of any of claims 35-37.

Images