Scheduling of machine learning related data

Abstract

A method includes receiving, at a first device from a second device, characteristic information of data to be transmitted to or from the second device, where the data is related to a machine learning model; and determining auxiliary information for scheduling transmission of data based at least on the characteristic information.

Description

Technical Field

[0001] Various example embodiments of the present disclosure relate generally to the field of telecommunications, and in particular, to methods, apparatus, and computer-readable storage media for scheduling machine learning (ML) related data. Background Art

[0002] Several technologies have been proposed to improve communication performance. For example, communication devices can adopt artificial intelligence (AI) / ML models to improve communication quality. AI / ML models can be applied to different scenarios to achieve better performance. To this end, AI / ML related data needs to be transmitted between different entities so that AI / ML models can be used. Summary of the invention

[0003] In a first aspect of the present disclosure, a first device is provided. The first device includes at least one processor; and at least one memory storing instructions, which, when executed by the at least one processor, cause the first device to at least: receive characteristic information of data to be sent to or from the second device from a second device, wherein the data is related to a machine learning model; and determine auxiliary information for scheduling transmission of the data based on at least the characteristic information.

[0004] In a second aspect of the present disclosure, a second device is provided. The second device includes at least one processor; and at least one memory storing instructions, which, when executed by the at least one processor, cause the second device to at least: send characteristic information of data to be sent to or from the second device to a first device, wherein the data is related to a machine learning model, and wherein the characteristic information is used to determine auxiliary information for scheduling transmission of the data.

[0005] In a third aspect of the present disclosure, a third device is provided. The third device includes at least one processor; and at least one memory storing instructions, which, when executed by the at least one processor, cause the third device to at least: receive auxiliary information for scheduling transmission of data to be sent to or from a second device from a first device, wherein the data is related to a machine learning model; determine scheduling information for or to be used for transmission of the data based on the auxiliary information; and send the scheduling information to the first device.

[0006] In a fourth aspect of the present disclosure, a method is provided. The method includes receiving, at a first device, characteristic information of data to be sent to or from the second device from a second device, wherein the data is related to a machine learning model; and determining auxiliary information for scheduling transmission of the data based at least on the characteristic information.

[0007] In a fifth aspect of the present disclosure, a method is provided. The method includes sending characteristic information of data to be sent to or from the second device from a second device to a first device, wherein the data is related to a machine learning model, and wherein the characteristic information is used to determine auxiliary information for scheduling transmission of the data.

[0008] In a sixth aspect of the present disclosure, a method is provided. The method includes receiving, at a third device, auxiliary information from a first device for scheduling transmission of data to be sent to or from a second device, wherein the data is related to a machine learning model; determining scheduling information for or to be used for transmission of the data based on the auxiliary information; and sending the scheduling information to the first device.

[0009] In a seventh aspect of the present disclosure, a first device is provided. The first device includes a component for receiving characteristic information of data to be sent to or from the second device from a second device, wherein the data is related to a machine learning model; and a component for determining auxiliary information for scheduling transmission of the data based at least on the characteristic information.

[0010] In an eighth aspect of the present disclosure, a second device is provided. The second device includes a component for sending characteristic information of data to be sent to or from the second device to a first device, wherein the data is related to a machine learning model, and wherein the characteristic information is used to determine auxiliary information for scheduling transmission of the data.

[0011] In a ninth aspect of the present disclosure, a third device is provided. The third device includes a component for receiving auxiliary information for scheduling transmission of data to be sent to or from a second device from a first device, wherein the data is related to a machine learning model; a component for determining scheduling information for or to be used for transmission of the data based on the auxiliary information; and a component for sending the scheduling information to the first device.

[0012] In a tenth aspect of the present disclosure, a computer-readable medium is provided, wherein the computer-readable medium includes instructions stored thereon, and the instructions are used to cause a device to at least execute the method according to the fourth aspect.

[0013] In an eleventh aspect of the present disclosure, a computer-readable medium is provided, wherein the computer-readable medium includes instructions stored thereon, and the instructions are used to cause a device to at least execute the method according to the fifth aspect.

[0014] In a twelfth aspect of the present disclosure, a computer-readable medium is provided, wherein the computer-readable medium includes instructions stored thereon, and the instructions are used to cause a device to at least execute the method according to the sixth aspect.

[0015] It should be understood that the invention summary is not intended to identify the key or essential features of the 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 readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Some example embodiments will now be described with reference to the accompanying drawings, in which:

[0017] Figure 1A An example communication environment is shown in which example embodiments of the present disclosure may be implemented;

[0018] Figure 1B Another example communication environment is shown in which example embodiments of the present disclosure may be implemented;

[0019] Figure 1C Another example communication environment is shown in which example embodiments of the present disclosure may be implemented;

[0020] Figure 2 shows an example signaling flow for scheduling ML-related data according to some example embodiments of the present disclosure;

[0021] Figure 3 shows an example signaling flow for scheduling ML-related data according to some example embodiments of the present disclosure;

[0022] Figure 4 A flowchart showing a method implemented at a first device according to some example embodiments of the present disclosure is shown;

[0023] Figure 5 A flowchart showing a method implemented at a second device according to some example embodiments of the present disclosure is shown;

[0024] Figure 6 A flowchart of a method implemented at a third device according to some example embodiments of the present disclosure is shown;

[0025] Figure 7 shows a simplified block diagram of a device suitable for implementing an example embodiment of the present disclosure; and

[0026] Figure 8 A block diagram of an example computer-readable medium is shown according to some example embodiments of the present disclosure.

[0027] Throughout the drawings, the same or similar reference numerals refer to the same or similar elements. DETAILED DESCRIPTION

[0028] The principles of the present disclosure will now be described with reference to some example embodiments. It should be understood that these embodiments are described only for illustrative purposes and help those skilled in the art to understand and implement the present disclosure without implying any limitation on the scope of the present disclosure. The embodiments described herein can be implemented in various ways except for the way described below.

[0029] 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 skill in the art to which this disclosure belongs.

[0030] References in this disclosure to "one embodiment," "an embodiment," "an example embodiment," etc. indicate that the described embodiment may include a particular feature, structure, or characteristic, but not every embodiment necessarily includes the particular feature, structure, or characteristic. Furthermore, such phrases do not necessarily refer to the same embodiment. Furthermore, when a particular feature, structure, or characteristic is described in conjunction with an embodiment, it is considered to be within the knowledge of those skilled in the art to affect such feature, structure, or characteristic in conjunction with other embodiments, whether or not explicitly described.

[0031] It should be understood that although the terms "first", "second", etc. in front of (multiple) nouns can be used in this article to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another element, and they do not limit the order of (multiple) nouns. For example, a first element can be referred to as a second element, and similarly, a second element can be referred to as a first element without departing from the scope of the exemplary embodiments. As used herein, the term "and / or" includes any and all combinations of one or more listed terms.

[0032] As used herein, “at least one of: ” and “at least one of ” and similar expressions, where a list of two or more elements is combined 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 of the elements.

[0033] As used herein, unless explicitly stated, performing a step "in response to A" does not indicate that the step is performed immediately after "A" occurs, and may include one or more intermediate steps.

[0034] The terms used herein are only used for the purpose of describing specific embodiments and are not intended to limit the exemplary embodiments. As used herein, the singular forms "a", "an", and "the" are also intended to include the plural forms unless the context clearly indicates otherwise. It will be further understood that the terms "comprise", "including", "having", "including", and / or "comprising" when used herein specify the presence of stated features, elements, and / or components, etc., but do not exclude the presence or addition of one or more other features, elements, components, and / or combinations thereof.

[0035] As used in this application, the term "circuitry" may refer to one, more, or all of the following: (a) pure hardware circuit implementation (such as pure analog and / or digital circuit system implementation), and (b) a combination of hardware circuitry and software such as (where applicable): (i) a combination of analog and / or digital hardware circuits and software / firmware, and (ii) any portion of hardware processor(s) (including digital signal processor(s)), software and memory(s) with software that work together to cause the device to (such as a mobile phone or a server) that performs various functions, and (c) (multiple) hardware circuits and / or (multiple) processors, such as (multiple) microprocessors or portions of (multiple) microprocessors, which require software (e.g., firmware) for operation, but the software may not be present when not required for operation.

[0036] This definition of circuitry applies to all uses of the term in this application, including any claims. As another example, as used in this application, the term circuitry also covers an implementation of only a hardware circuit or processor (or multiple processors) or a portion of a hardware circuit or processor and its (or its) 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 a server, cellular network device, or other computing or network device.

[0037] As used herein, the term "communication network" refers to a network that follows any suitable communication standard, such as New Radio (NR), Long Term Evolution (LTE), Advanced LTE (LTE-A), Wideband Code Division Multiple Access (WCDMA), High Speed ​​Packet Access (HSPA), Narrowband Internet of Things (NB-IoT), etc. In addition, the communication between the terminal equipment and the network equipment in the communication network can be performed according to any suitable generation of communication protocols, including but not limited to the first generation (1G), second generation (2G), 2.5G, 2.75G, third generation (3G), fourth generation (4G), 4.5G, fifth generation (5G), sixth generation (6G) communication protocols and / or any other protocols currently known or developed in the future. The embodiments of the present disclosure can be applied to various communication systems. In view of the rapid development of communication, there will certainly be communication technologies and systems that can embody future types of the present disclosure. It should not be considered that the scope of the present disclosure is limited to the aforementioned system.

[0038] 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 from it. A network device may refer to a base station (BS) or an access point (AP), for example, a Node B (Node B or NB), an evolved NodeB (eNode B or eNB), a NR NB (also referred to as a gNB), a remote radio unit (RRU), a radio head (RH), a remote radio head (RRH), a relay, an integrated access and backhaul (IAB) node, a low-power node such as a femto, a micro, a non-terrestrial network (NTN) or a non-terrestrial 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, etc., depending on the terminology and technology applied. In some example embodiments, a radio access network (RAN) split architecture includes a centralized unit (CU) and a distributed unit (DU) at an IAB donor node. An IAB node includes a mobile terminal (IAB-MT) portion that behaves as a UE to a parent node, and a DU portion of the IAB node behaves as a base station to a next-hop IAB node.

[0039] The term "terminal device" refers to any terminal device capable of wireless communication. As an example and not limitation, a terminal device may also be referred to as a communication device, a 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 is not limited to, a mobile phone, a cellular phone, a smart phone, a voice over IP (VoIP) phone, a wireless local loop phone, a tablet computer, a wearable terminal device, a personal digital assistant (PDA), a portable computer, a desktop computer, an image capture terminal device such as a digital camera, a game terminal device, a music storage and playback device, a vehicle-mounted wireless terminal device, a wireless endpoint, a mobile station, a laptop embedded device (LEE), a laptop mounted device (LME), a USB dongle, a smart device, a wireless customer equipment (CPE), an Internet of Things (IoT) device, a watch or other wearable device, a head mounted display (HMD), a vehicle, a drone, medical equipment and applications (e.g., remote surgery), industrial equipment and applications (e.g., robots and / or other wireless devices operating in the context of an industrial and / or automated processing chain), consumer electronic devices, equipment operating on a commercial and / or industrial wireless network, etc. The terminal device may also correspond to the mobile terminal (MT) part of an IAB node (eg, a relay node).In the following description, the terms "terminal device", "communication device", "terminal", "user equipment" and "UE" may be used interchangeably.

[0040] As used herein, the terms "resource", "transmission resource", "resource block", "physical resource block" (PRB), "uplink resource" or "downlink resource" may refer to any resource used to perform communication, for example, communication between a terminal device and a network device, such as time domain resources, frequency domain resources, spatial domain resources, code domain resources, or any other combination of time domain, frequency domain, spatial and / or code domain resources that enable communication. In the following, unless explicitly stated, resources in the frequency domain and the time domain will be used as examples of transmission resources to describe some example embodiments of the present disclosure. Note that the example embodiments of the present disclosure are equally applicable to other resources in other domains.

[0041] 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," "AI model," and "ML model." The term "AI / ML" may be used interchangeably with the terms "AI" and "ML."

[0042] As used herein, the term "data collection" may refer to the process of collecting data by a network node, a management entity, or a UE for any suitable purpose. The purpose may include, but is not limited to, training, data analysis and reasoning, monitoring (e.g., performance monitoring), retraining, updating, and tuning AI / ML models / functions / features.

[0043] As used herein, the term "data associated with an ML model" or "ML-related data" may refer to data used for any suitable purpose of an ML model or used for any suitable stage associated with an ML model. For example, ML-related data may include data used for one or more of ML model training, monitoring, retraining, analysis, and reasoning. Hereinafter, the terms "data associated with an ML model" and "ML-related data" may be used interchangeably.

[0044] The term "UE-side (AI / ML) model" used herein may refer to an AI / ML model that performs inference entirely at the UE. The term "network-side (AI / ML) model" used herein may refer to an AI / ML model that performs inference entirely at the network. The term "single-sided (AI / ML) model" used herein may refer to a UE-side (AI / ML) model or a network-side (AI / ML) model. The term "double-sided (AI / ML) model" used herein may refer to paired (multiple) AI / ML models on which joint inference is performed, wherein the joint inference includes AI / ML inference that performs inference jointly across the UE and the network, i.e., the first part of the inference is first performed by the UE, and then the remaining part is performed by the gNB, or vice versa.

[0045] Figure 1A , Figure 1B and Figure 1C Example communication environments 100A, 110B, and 100C are shown in which example embodiments of the present disclosure may be implemented. In any of the communication environments 100A, 100B, and 100C, machine learning (e.g., one or more ML models) may be used for functions including, but not limited to, positioning, beam management, and channel state information (CSI) feedback. Such functions may also be referred to as AI / ML-related functions or ML-related functions.

[0046] Specifically, environment 100A includes a terminal device 120-1, a radio access network (RAN) device 131, and a core network (CN) device 111. The terminal device 110-2 can be any suitable type of terminal device, such as a UE or a positioning reference unit (PRU). The RAN device 131 can be any suitable RAN node, such as a gNB. The CN device 111 may include one or more network functions (NFs). The specific NF may depend on the AI / ML function. For example, in the case of AI / ML for positioning, the CN device 111 may include a location management function (LMF) and / or a network data analysis function (NWDAF). Alternatively, the LMF and / or NWDAF may be implemented at a RAN device, which may be different from or the same as the RAN device 131.

[0047] The LMF may request positioning-related capabilities of the UE, provide assistance data to the UE (e.g., configuration of reference nodes and positioning signals), and obtain location information (location measurement data and / or location estimate from the target UE). The LMF may communicate with the target device (i.e., UE) using the LTE Positioning Protocol (LPP) and communicate with the gNB using the New Radio Positioning Protocol A (NRPPa) protocol.

[0048] The LMF may decide on the positioning method (e.g. based on QoS, UE / gNB capabilities) and invoke it at the UE / gNB, either via UE-based positioning (where the UE computes a position estimate) or via UE-assisted / NW-based positioning (where the UE provides measurements and the LMF computes a position estimate). The LMF may combine all results and determine a single position estimate as well as the accuracy of the estimate and velocity.

[0049] A PRU is a UE with a known location, supporting UL / DL positioning measurements, which can be collected by the LMF and compared with the known PRU location to determine correction terms for other nearby UEs.

[0050] NWDAF can collect location information for a target UE or a group of target UEs from the LCS system. The collected location-related information may include, but is not limited to, the UE's location estimate; the timestamp of the location estimate; the UE's speed; information about the positioning method used to obtain the UE's location estimate; indications of area events when the UE enters, is in, or leaves a geographic area; indications of motion events when the UE moves more than a predefined straight-line distance from a previous location.

[0051] NWDAF may consider checking user consent for the purpose of data collection and use of such data. NWDAF checks user consent by obtaining user consent information from UDM.

[0052] User consent is subscription information stored in the Unified Data Management (UDM) function, which can include whether the user authorizes the collection and use of their data for a specific purpose; and the purpose of data collection, such as analysis or model training.

[0053] Environment 100B includes terminal device 120-2, RAN device 132, and terminal device 112. In this example, terminal device 120-2 can perform sidelink (SL) communication and operate in SL resource allocation mode 1, which is also called a network scheduling resource allocation scheme. In SL resource allocation mode 1, resources for sidelink communication by terminal device 120-2 can be scheduled by RAN device 132. RAN device 132 can be any suitable RAN node, such as a gNB. Terminal device 112 is different from terminal device 120-2. For example, terminal device 112 can be a terminal device that performs SL communication with terminal device 120-2, or a server terminal device, such as a server UE.

[0054] Environment 100C includes terminal device 120-3 and terminal device 113. In this example, terminal device 120-3 can perform SL communication and operate in SL resource allocation mode 2, which can also be referred to as an autonomous resource selection scheme. In SL resource allocation mode 2, resources for sidelink communication are selected by terminal device 120-3. In other words, terminal device 120-3 in SL resource allocation mode 2 can schedule SL transmission by itself without being scheduled by the network. Similar to Figure 1B The terminal device 113 may be a terminal device that performs SL communication with the terminal device 120 - 3 , or a server terminal device, such as a server UE.

[0055] The communication in the communication environments 100A, 100B and 100C may be implemented according to any appropriate (multiple) communication protocols, including but not limited to cellular communication protocols of the first generation (1G), second generation (2G), third generation (3G), fourth generation (4G), fifth generation (5G), sixth generation (6G), etc., wireless local area network communication protocols such as Institute of Electrical and Electronics Engineers (IEEE) 802.11, etc., and / or any other protocol currently known or developed in the future. In addition, the communication may utilize any appropriate wireless communication technology, including but not limited to: code division multiple access (CDMA), frequency division multiple access (FDMA), time division multiple access (TDMA), frequency division duplex (FDD), time division duplex (TDD), multiple input multiple output (MIMO), orthogonal frequency division multiple access (OFDM), discrete Fourier transform spread OFDM (DFT-s-OFDM) and / or any other technology currently known or developed in the future.

[0056] Hereinafter, terminal devices 120-1, 120-2, and 120-3 may be collectively referred to as terminal devices 120 or individually referred to as terminal devices 120. As described above, in any of the communication environments 100A, 100B, and 100C, ML-related functions may be used. In order to enable ML-related functions, data collection is required. For example, the terminal device 120 may have ML-related data to be sent to another entity, or another entity may have ML-related data to be sent to the terminal device 120. Now, taking AI / ML for positioning as an example, some examples for data collection are described.

[0057] Use cases for positioning can include a one-step approach (also known as direct AI / ML positioning) and a two-step approach (also known as AI / ML assisted positioning). In the one-step approach, the AI / ML model outputs the UE location, for example, fingerprint recognition based on channel observations as input to the AI / ML model. In the two-step approach, the AI / ML model outputs new measurements and / or enhancements to existing measurements, for example, line of sight (LOS) / non-line of sight (NLOS) identification, timing and / or angle of measurement, and likelihood of measurement.

[0058] More specifically, the following situations are possible: - Case 1: UE-based positioning with UE-side model, direct AI / ML or AI / ML-assisted positioning; - Case 2a: UE-assisted / LMF based positioning with UE-side model, AI / ML-assisted positioning; - Case 2b: UE-assisted / LMF based positioning with LMF side model, direct AI / ML positioning; - Case 3a: Next Generation (NG)-RAN node assisted positioning with gNB side model, AI / ML assisted positioning; - Case 3b: NG-RAN node assisted positioning with LMF side model, direct AI / ML positioning.

[0059] One-sided models where inference is performed entirely at the UE or the network may be prioritized.

[0060] Examples are now given of some aspects of signaling, reporting, and feedback of data collection for ML-based positioning. One aspect may include details of a request / report of labels and / or other training data, and enable delivery of collected labels and / or other training data to a training entity when the training entity is not the same entity that acquired the labels and / or other training data.

[0061] Another aspect may include auxiliary signaling indicating reference signal configuration(s) to derive signatures and / or other training data.

[0062] Another aspect may include a request / report of training data, such as but not limited to: true labels; measurements corresponding to model inputs; associated information for true labels and / or measurements corresponding to model inputs.

[0063] Another aspect may include auxiliary signaling and processes to support the generation of training data, such as but not limited to: (multiple) reference signal (e.g., positioning reference signal (PRS) / sounding reference signal (SRS)) configuration and configuration identifier; auxiliary information, for example, between LMF and UE / PRU, for tag calculation / generation, and tag validity / quality conditions, etc.

[0064] It should be noted that whether such auxiliary signaling and processes can be applied to (multiple) other aspects of AI / ML model lifecycle management (LCM) will be specified. It should also be noted that different entities can be considered to generate training data, and different types of training data can be considered when applicable, and the following two cases can be considered when applicable: when the training entity and the entity generating the training data are the same entity, and when the training entity is not the same entity as the entity generating the training data.

[0065] Now give some examples of aspects of training data collection for AI / ML-based positioning. The associated information of the training data may include, but is not limited to, at least true labels for model training; reports from label data generating entities; at least measurements for model training (corresponding to model inputs); reports from measurement data generating entities; quality indicators for and / or associated with true labels and / or measurements for at least model training; reports from label and / or measurement data generating entities and / or reports requested from different (e.g., data collection, etc.) entities; at least (multiple) RS configurations for derived measurements; at least training data for model training and / or timestamps associated with the training data for model training; reports from data generating entities and training data and / or reports as LMF auxiliary signaling. RS configuration may involve a request from a data generating entity (e.g., UE / PRU / transmit and receive point (TRP)) to LMF and / or to UE / PRU / TRP as LMF auxiliary signaling. When the measurements and true labels are generated by different entities, the timestamps may include separate timestamps for the measurements and true labels.

[0066] Auxiliary signaling and processes used to support the generation / collection of training data may include, but are not limited to, potential determination of the UE / PRU / TRP that can provide training data; configuration of reference signals (for measurement and / or marking); signaling other than the above two items for data collection, for example, the requested training data quality.

[0067] In addition to training, data collection aspects have been identified for model monitoring purposes. These data collection aspects may include, but are not limited to: if certain types of data are necessary to calculate monitoring metrics, how entities may be used to provide data of a given type to calculate monitoring metrics; potential signaling for providing data of a given type that is used to calculate associated monitoring metrics; and potential auxiliary signaling and processes for supporting entities in providing data for calculating monitoring metrics.

[0068] Regarding data collection, some solutions may focus on data collection solutions for both NW-side and UE-side AI / ML models, including auxiliary signaling and (dataset) reporting from relevant entities. The impact of data collection for all relevant LCM purposes, such as model training / monitoring / selection / update / inference, etc., may be studied. In some solutions, data collection requirements and solutions for different LCM purposes may be analyzed separately. In some solutions, the following metrics may be considered as starting points: a) content of the data, b) size of the data, c) latency and periodicity, d) signaling, entities involved, and configuration aspects.

[0069] Several example aspects regarding data collection for AL / ML models are described. Typically, ML-based positioning methods require large amounts of data, primarily for training, monitoring, and retraining ML models. The data includes positioning-related measurements that are typically input to the model, such as time / angle / phase / power based measurements made by the UE using DL PRS or by the gNB using UL SRS, channel impulse response (CIR), packet data protocol (PDP), and any real information associated with them (such as UE location information, LOS / NLOS channel type indication, etc.), which are used as "labels" for training supervised ML models.

[0070] If other ML-based methods such as beam management and channel state information feedback are considered in addition to positioning, the amount of data that needs to be sent for ML-related purposes may actually be larger. In other words, a large amount of data is expected to be sent between the UE and the network entity in the upcoming release to address multiple ML-assisted use cases. In the following, some example embodiments are described taking the use case of positioning as an example.

[0071] According to some mechanisms, user data to be transmitted between entities may be prioritized based on data plane quality of service (QoS) priorities without regard to ML-related data transmission. That is, the decision of which user data packets are transmitted first does not take into account potential data for ML purposes, which, as described above, may occupy a large amount of data on the order of application layer data.

[0072] In these mechanisms, ML-related priorities are ignored in data transmission scheduling. The impact of this ignoring of ML-related priorities in scheduling data transmission is that ML-based solutions for air interfaces (such as positioning, for example) cannot work properly. Therefore, there is a need for an effective mechanism to incorporate data for ML-based air interface use cases into QoS prioritization and / or scheduling of data transmission processes.

[0073] According to some example embodiments of the present disclosure, a scheme for scheduling ML-related data is provided. In the scheme, a first entity may receive characteristic information of ML-related data to be sent to the second entity or characteristic information of ML-related data from the second entity from a second entity. Based on the received characteristic information, the first entity may determine auxiliary information for scheduling the transmission of the ML-related data. The auxiliary information may then be used by an entity that schedules data transmission for the second entity to determine scheduling information for the transmission of the ML-related data. The ML-related data may be sent according to the scheduling information. Alternatively, the ML-related function may be updated according to the scheduling information.

[0074] In the proposed scheme, the characteristics of ML-related data are taken into account for scheduling the transmission of ML-related data. In this way, ML-related functions, such as positioning, beam management, and CSI feedback, can work properly, thereby improving communication performance.

[0075] Hereinafter, exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings.

[0076] Figure 2 An example signaling flow 200 for scheduling ML-related data according to some example embodiments of the present disclosure is shown. The signaling flow 200 may involve a first device 201, a second device 202, and a third device 203. The ML model may be included or implemented in any one or more of the first device 201, the second device 202, and the third device 203. For example, the ML model may be run or executed at one or more of the first device 201, the second device 202, and the third device 203 for any suitable purpose, such as training, retraining, updating, reasoning, monitoring. Data transmission for the ML model is required. For example, data related to the ML model (also referred to as ML-related data) will be sent to or from the second device 202.

[0077] ML-related data may be used for any suitable purpose, e.g., training monitoring, retraining of ML models. Functions or use cases of ML models may include, e.g., positioning, CSI feedback, and beam management. Other functions or use cases are also possible.

[0078] In process 200, third device 203 may schedule data transmission for second device 202. In some example embodiments, process 200 may be implemented in environment 100A. Accordingly, first device 201 may be implemented as CN device 111 or at CN device 111, second device 202 may be implemented as terminal device 120-1 or at terminal device 120-1, and third device 203 may be implemented as RAN device 131 or at RAN device 131. For example, in case of ML-related positioning, first device 201 may be implemented as LMF or NWDAF or at LMF or NWDAF, second device 202 may be implemented as UE or PRU or at UE or PRU, and third device 203 may be implemented as gNB or at gNB.

[0079] In some example embodiments, the process 200 may be implemented in the environment 100B. In such an example embodiment, the transmission of the ML-related data may be an SL transmission. Accordingly, the first device 201 may be implemented as the terminal device 112 or at the terminal device 112, the second device 202 may be implemented as the terminal device 120-2 or at the terminal device 120-2, and the third device 203 may be implemented as the RAN device 132 or at the RAN device 132. For example, the first device 201 may be implemented as a server UE or at a server UE, the second device 202 may be implemented as a UE operating in SL resource allocation mode 1 or at a UE operating in SL resource allocation mode 1, and the third device 203 may be implemented as a gNB serving the UE or at a gNB serving the UE.

[0080] like Figure 2 As shown, the second device 202 may send (210) characteristic information of ML-related data to the first device 201. For example, in the case of positioning, a UE having an ML-assisted positioning session with an LMF may provide characteristics of ML-related data to the LMF.

[0081] The characteristic information may include or indicate any suitable characteristic of the ML-related data. For example, the characteristic information may indicate the amount of the ML-related data. In another example, for example, in the case of a positioning use case, if the data is to be used for one-step or two-step positioning, the characteristic information may include the data type of the ML-related data.

[0082] In some example embodiments, the characteristic information may relate to one or more of a location, a timestamp, a configuration associated with the ML-related data or a measurement used to obtain the ML-related data. The configuration may include a TRP / anchor point, a PRS resource, a bandwidth (BW), a comb-size, etc. Alternatively or additionally, the characteristic information may also relate to statistics of the ML-related data, such as a minimum value, a maximum value, a range of values, an average value, a variance of values, etc.

[0083] In some example embodiments, Figure 2 As shown, the first device 201 can send (205) a request for characteristic information to the second device 202. For example, the request can be included in an LPP message. Then, in response to the request, the second device 202 can send (210) the requested characteristic information to the first device 201.

[0084] Continuing with process 200, first device 201 may receive characteristic information from second device 202, and may determine (215) assistance information for scheduling transmission of ML-related data based at least on the characteristic information. In some example embodiments, the assistance information may be further determined based on QoS information (e.g., QoS requirements) for ML-related functions. For example, in the case of ML-related positioning, the assistance information may be further determined based on positioning QoS (such as positioning accuracy and positioning latency).

[0085] The assistance information may include a priority of the ML-related data, also referred to as "data priority". Alternatively or additionally, the assistance information may also include or indicate one or more scheduling requirements for sending the ML-related data. Example scheduling requirements may include, but are not limited to, QoS boundaries, ranges (e.g., maximum / minimum / range of throughput), data volume, payload, latency, modulation and coding scheme (MCS), etc.

[0086] The first device 201 may send (220) assistance information to the third device 203. For example, in the case of ML-related positioning, the CN device 111 may initiate a new information element (IE) as part of the positioning RAN procedure to indicate the data priority of the ML-related data to the serving RAN device 131 (e.g., serving gNB) of the terminal device 110-1.

[0087] The third device 203 may receive the auxiliary information from the first device 201 and determine (225) scheduling information for transmission of the ML-related data based on the auxiliary information. The scheduling information may include a scheduling priority for the transmission of the ML-related data. Alternatively or additionally, the scheduling information may include one or more parameters related to the QoS achieved or achievable by transmitting the ML-related data. The parameters related to the QoS may include, for example, but not limited to, throughput, latency, etc., as well as those parameters described with respect to the auxiliary information.

[0088] In some example embodiments, third device 203 may determine scheduling information based on the assistance information and the user plane status of second device 202. For example, the scheduling priority derived by third device 203 may be a function of the data priority provided in the assistance information combined with the data plane function of second device 202.

[0089] The third device 203 may send (230) scheduling information to the first device 201. In some example embodiments, the third device 203 may schedule the transmission of the ML-related data based on the scheduling information. In such example embodiments, the scheduling information received at the first device 201 has been used for the transmission of the ML-related data. Therefore, the scheduling information indicates the actual or realized scheduling of the ML-related data. In other words, the third device 203 may perform the actual scheduling and notify the first device 201 of the scheduling information used in the actual scheduling.

[0090] Alternatively, in some example embodiments, the third device 203 may send scheduling information to the first device 201 before performing the actual scheduling. In such example embodiments, the scheduling information received at the first device 201 will be used for the transmission of the ML-related data. The scheduling information indicates an achievable schedule for the ML-related data.

[0091] The first device 201 may receive scheduling information for or to be used for the transmission of ML-related data from the third device 203. The first device 201 may then evaluate (235) or assess the scheduling information for the QoS requirements of the ML-related functions. In other words, the first device 201 may determine whether the scheduling information meets the QoS requirements for the ML-related functions. In the case of ML-related positioning, the scheduling information may be evaluated for positioning QoS. For example, according to the QoS, the maximum delay for data transmission may be AA ms, while the scheduling information indicates a delay of BB ms for transmitting ML-related data. If BB does not exceed AA, the QoS requirements are met. If BB exceeds AA, the QoS requirements may not be met.

[0092] If the first device 201 determines that the QoS requirement is not met, process 280 may be performed. Specifically, the first device 201 may determine (240) to update the ML-related functions. In other words, the first device 201 may adjust the ML method for the function under consideration (e.g., positioning, CSI feedback, or beam management). The updated method may require a smaller amount of data, or may be more robust to the ML-related data to be sent. It should be understood that in the case where the scheduling information received from the third device 203 has been used for scheduling, if the QoS requirement is not met, the ML-related functions may also be updated, which means that process 280 may be performed.

[0093] An update to an ML-related function may include a switch from an ML model to another ML model designed for the function, such as a switch from a first ML model used in a one-step approach to a second ML model used in a two-step approach. Alternatively or additionally, the update may include a fallback from an ML-based approach to a conventional approach. Other types of updates are also possible.

[0094] As an example and not any limitation, in the case of ML-related positioning, the LMF may use or select a lower rank positioning method that requires a smaller amount of data or is more robust to the ML-related data provided by the UE. For example, a two-step method that uses an ML model only for intermediate parameters of the final position estimate may be selected instead of a one-step method (e.g., fingerprinting) that depends only on ML-related data to derive a position estimate.

[0095] Next, the first device 201 may send (245) an indication to update the ML-related function to the second device 202 and / or the third device 203. In the environment 100A, the indication may be sent from the CN device 111 to the terminal device 110-2 via LPP, and to the RAN device 131 via NRPPa. In the environment 100B, the indication may be sent from the terminal device 112 to the terminal device 120-1 via sidelink signaling, and to the RAN device 132 via any suitable uplink signaling.

[0096] Accordingly, the second device 202 and the third device 203 may receive an instruction to update the ML-related function from the first device 201. Then, the second device 202 and the third device 203 may update the ML-related function accordingly.

[0097] Now return to reference action 235. If the first device 201 determines that the QoS requirement is met, process 290 can be performed. Specifically, the first device 201 can send a confirmation of the scheduling information to the third device 203. Accordingly, the third device 203 can receive the confirmation from the first device 201. If the scheduling information sent at 230 has been used for scheduling, the third device 203 can continue to use the scheduling information for subsequent transmission of ML-related data. If the scheduling sent at 230 will be used for scheduling, the third device 203 can schedule (255) the transmission of ML-related data. The second device 202 can receive (260) a scheduling indication from the third device 203 and perform (265) data transmission based on the scheduling indication.

[0098] For example, in environment 100A, RAN device 131 (such as gNB) can schedule terminal device 120-1 for UL / DL transmission according to scheduling information. Data transmission can also involve CN device 111 (e.g., LMF other than gNB) as a source, destination or hop. For another example, in environment 100B, RAN device 132 (such as gNB) can schedule terminal device 120-2 for SL transmission according to scheduling information. Data transmission can be performed between terminal device 120-2 and terminal device 112 and / or between terminal device 120-2 and another terminal device (not shown).

[0099] Figure 3 Another example signaling flow 300 for scheduling ML-related data according to some example embodiments of the present disclosure is shown. The signaling flow 300 may involve a first device 201 and a second device 202. The ML model may be included in any one or more of the first device 201, the second device 202, and another device (not shown). Data transmission for the ML model is required. For example, data related to the ML model (also referred to as ML-related data) is to be sent to or from the second device 202. The ML-related data may be the same as that described above with reference to Figure 2 The data described are similar, so their description will not be repeated here.

[0100] In process 300, second device 202 may schedule data transmission by itself. For example, second device 202 may include a UE in SL resource allocation mode 2. In some example embodiments, process 300 may be implemented in environment 100C. Accordingly, first device 201 may be implemented as terminal device 113 or at terminal device 113, and second device 202 may be implemented as terminal device 120-3 or at terminal device 120-3. For example, in the case of ML-related positioning, first device 201 may be implemented as a server UE or at a server UE, and second device 202 may be implemented as a UE or a PRU or at a UE or a PRU.

[0101] like Figure 3 As shown, the second device 202 may send (310) characteristic information of ML-related data to the first device 201. For example, in the case of positioning, a UE having an ML-assisted positioning session with a server UE may provide the characteristics of the ML-related data to the server UE. The characteristic information may be the same as that in the above reference Figure 2 The characteristic information described is similar, so its description will not be repeated here.

[0102] In some example embodiments, Figure 3 As shown, the first device 201 may send (305) a request for characteristic information to the second device 202. Then, in response to the request, the second device 202 may send (310) the requested characteristic information to the first device 201.

[0103] Continuing with process 300, first device 201 may receive characteristic information from second device 202, and may determine (215) assistance information for scheduling transmission of ML-related data based at least on the characteristic information. In some example embodiments, the assistance information may be further determined based on QoS information (e.g., QoS requirements) for ML-related functions. For example, in the case of ML-related positioning, the assistance information may be further determined based on positioning QoS (such as positioning accuracy and positioning latency).

[0104] The assistance information may include a priority of the ML-related data, which is also referred to herein as "data priority". Alternatively or additionally, the assistance information may also include or indicate one or more scheduling requirements for sending the ML-related data. Example scheduling requirements may include, but are not limited to, QoS boundaries, ranges (e.g., maximum / minimum / range of throughput), data volume, payload, latency, modulation and coding scheme (MCS), etc.

[0105] The first device 201 may send (320) the auxiliary information to the second device 202. Accordingly, the second device 202 may receive the auxiliary information from the first device 201 and determine (325) scheduling information for transmission of the ML-related data based on the auxiliary information. The scheduling information may include a scheduling priority for the transmission of the ML-related data. Alternatively or additionally, the scheduling information may include one or more parameters related to the QoS achieved or achievable for transmitting the ML-related data. The parameters related to the QoS may include, for example, but not limited to, throughput, latency, etc., as well as those parameters described with respect to the auxiliary information.

[0106] In some example embodiments, second device 202 may determine scheduling information based on the assistance information and a user plane status of second device 202. For example, the scheduling priority derived by second device 202 may be a function of the data priority provided in the assistance information combined with the data plane function of second device 202.

[0107] The second device 202 may send (330) scheduling information to the first device 201. In some example embodiments, the third device 203 may schedule the transmission of the ML-related data based on the scheduling information. The transmission of the ML-related data may be performed based on the scheduling information. In such an example embodiment, the scheduling information received at the first device 201 has been used for the transmission of the ML-related data. Therefore, the scheduling information indicates the actual or realized scheduling of the ML-related data. In other words, the second device 202 may perform the actual scheduling and notify the first device 201 of the scheduling information used in the actual scheduling.

[0108] Alternatively, in some example embodiments, the second device 202 may send scheduling information to the first device 201 before performing the actual scheduling. In such example embodiments, the scheduling information received at the first device 201 will be used for the transmission of the ML-related data. The scheduling information indicates an achievable schedule for the ML-related data.

[0109] The first device 201 may receive scheduling information for or to be used for transmission of ML-related data from the second device 202. The first device 201 may then evaluate (335) or assess the scheduling information against QoS requirements of ML-related functions. In other words, the first device 201 may determine whether the scheduling information meets the QoS requirements for ML-related functions. In the case of ML-related positioning, the scheduling information may be evaluated against positioning QoS.

[0110] If the first device 201 determines that the QoS requirement is not met, process 380 may be performed. Specifically, the first device 201 may determine (340) to update the ML-related functions. In other words, the first device 201 may adjust the ML method for the function under consideration (e.g., positioning, CSI feedback, or beam management). The updated method may require a smaller amount of data or may be more robust to the ML-related data to be sent. The update of the ML-related functions is similar to the above reference Figure 2 The described updates are similar and therefore not repeated here. It should be understood that in the case where the scheduling information received from the first device 201 has been used for scheduling, if the QoS requirements are not met, the ML related functions may also be updated, which means that process 380 may be performed.

[0111] As an example and not any limitation, in the case of ML-related positioning, the server UE may use or select a lower rank positioning method that requires a smaller amount of data or is more robust to the ML-related data provided by the UE. For example, a two-step method that uses an ML model only for intermediate parameters of the final position estimate may be selected instead of a one-step method that depends only on ML-related data to derive a position estimate.

[0112] Next, the first device 201 may send (345) an instruction to update the ML-related function to the second device 202. Accordingly, the second device 202 may receive the instruction to update the ML-related function from the first device 201 and may then update the ML-related function accordingly.

[0113] Now return to reference action 335. If the first device 201 determines that the QoS requirement is met, process 390 can be performed. Specifically, the first device 201 can send a confirmation of the scheduling information to the second device 202. Accordingly, the second device 202 can receive the confirmation from the first device 201. If the scheduling information sent at 330 has been used for scheduling, the second device 202 can continue to use the scheduling information for subsequent transmission of ML-related data. If the schedule sent at 330 will be used for scheduling, the second device 202 can schedule (355) the transmission of ML-related data, and can perform (365) data transmission via the side link accordingly.

[0114] For example, in the environment 100C, the terminal device 120-3 may perform SL data transmission of ML related data according to the confirmed scheduling information. The data transmission may be performed between the terminal device 120-3 and the terminal device 113 and / or between the terminal device 120-3 and another terminal device (not shown).

[0115] Some example embodiments are described above. The present disclosure provides a method for adjusting the scheduling of data plane transmissions of a UE based on the content of data to be used for ML-related purposes. In order to better understand the proposed scheme for scheduling ML-related data transmission, an example process is now described. A network entity associated with the ML-related method determines auxiliary information for scheduling the transmission of ML-related data, for example, a priority for the ML-related data. For example, in an ML-assisted positioning use case, the LMF determines the priority of data to be sent between the UE and the network for ML-assisted positioning. For simplicity, the following description focuses on the positioning case, so that the network entity of interest is the LMF, however the example can be generalized for other use cases of interest (such as beam management or CSI feedback).

[0116] As part of the example process, the UE provides the LMF with characteristics of the ML-related data to be sent or received. These characteristics may include, for example, the amount of data or the type of data - for positioning use cases, this may be one-step positioning data or two-step positioning data, etc. The LMF then determines the assistance information for the ML-related data based on the characteristics provided by the UE in the previous step, for example, the priority of such data for the purpose of ML-related functions (e.g., positioning, beam management, or CSI feedback).

[0117] For the UE of interest, the LMF indicates the assistance information it acquired in the previous step to the corresponding RAN node (e.g., the serving gNB(s) of the corresponding UE). For example, the data priority may be indicated to the corresponding RAN node and the priority may be associated with the transmission of data for positioning purposes (e.g., for training a positioning model). The RAN node (e.g., gNB) derives scheduling information (such as scheduling priority) based on the data plane details of the UE (which are sensed at the gNB and are independent of the positioning process) in combination with the ML-related data priority provided by the LMF. The RAN node may schedule the UE accordingly and indicate the scheduling information to the LMF.

[0118] The LMF then evaluates the scheduling information provided by the gNB. If the scheduling information does not meet the positioning QoS requirements (known at the LMF), the LMF adjusts the positioning method. As an example, the LMF may select an alternative ML-assisted positioning method that requires a smaller amount of data to accommodate the positioning method on the available scheduling priority provided by the gNB. For example, the LMF may switch from a one-step method that requires a large amount of data to a two-step method that requires a relatively small amount of data. In another example, in the case where the scheduling information from the gNB is not sufficient to support ML-assisted positioning, the LMF may decide to switch from ML-assisted positioning to non-ML positioning. The LMF then uses LPP to update the UE with the newly selected positioning method and / or positioning mode.

[0119] Example flows and processes are described above. In this way, an efficient mechanism for scheduling data transmission for AI / ML can be implemented.

[0120] Figure 4 FIG. 4 is a flowchart of an example method 400 implemented at a first device according to some example embodiments of the present disclosure. For discussion purposes, Figure 2 or Figure 3 The method 400 is described from the perspective of the first device 201 in FIG.

[0121] At block 410, a first device receives characteristic information of data to be sent to or from the second device from a second device. The data is related to a machine learning model.

[0122] At block 420, the first device determines assistance information for scheduling transmission of data based at least on the characteristic information.

[0123] In some example embodiments, the first device may send assistance information to a third device, the third device scheduling data transmission for the first device; and receive scheduling information for or to be used for transmission of data from the third device.

[0124] In some example embodiments, method 400 also includes determining whether the scheduling information meets the quality of service requirements for the functionality of the machine learning model; and if it is determined that the scheduling information meets the quality of service requirements, sending a confirmation of the scheduling information to a third device.

[0125] In some example embodiments, method 400 further includes: if it is determined that the scheduling information does not meet the quality of service requirements, determining an update to the functionality of the machine learning model based on the scheduling information; and sending an indication for updating the functionality of the machine learning model to at least one of the third device or the second device.

[0126] In some exemplary embodiments, the first device includes a core network device or a radio access network device, the second device includes a terminal device, the third device includes a radio access network device, and the transmission of data includes at least one of the following: uplink transmission from the terminal device to the radio access network device, or downlink transmission from the radio access network device to the terminal device.

[0127] In some exemplary embodiments, the first device includes a first terminal device, the second device includes a second terminal device, the third device includes a radio access network device, and the transmission of data includes at least one of the following: a side link transmission between the first terminal device and the second terminal device, or a side link transmission between the second terminal device and the third terminal device.

[0128] In some example embodiments, the first apparatus may transmit assistance information to the second apparatus; and receive scheduling information used or to be used for transmission of data from the second apparatus.

[0129] In some example embodiments, method 400 further includes: determining whether the scheduling information satisfies the quality of service requirements for the functionality of the machine learning model; and if it is determined that the scheduling information satisfies the quality of service requirements, sending a confirmation of the scheduling information to the second device.

[0130] In some example embodiments, method 400 further includes: if it is determined that the scheduling information does not meet the quality of service requirements, determining an update to the functionality of the machine learning model based on the scheduling information; and sending an indication to the second device for updating the functionality of the machine learning model.

[0131] In some example embodiments, the first apparatus comprises a terminal device and the second apparatus comprises another terminal device.

[0132] In some example embodiments, the method 400 further includes sending a request for characteristic information of the data to the second device.

[0133] In some example embodiments, the machine learning model is implemented in at least one of: the first device, the second device, or a third device for scheduling data transmission to the second device.

[0134] Figure 5 FIG. 5 is a flowchart of an example method 500 implemented at a second device according to some example embodiments of the present disclosure. For discussion purposes, Figure 2 or Figure 3 The method 500 is described from the perspective of the second device 202.

[0135] At block 510, the second device sends characteristic information of data to be sent to or from the second device to the first device, wherein the data is related to the machine learning model. The characteristic information is used to determine auxiliary information for scheduling transmission of the data.

[0136] In some example embodiments, method 500 further includes receiving, from the first device, an indication of functionality for updating the machine learning model.

[0137] In some example embodiments, the method 500 further includes: receiving scheduling information for transmission of the data from the third device; and performing the transmission based on the scheduling information.

[0138] In some exemplary embodiments, the first device includes a core network device or a radio access network device, the second device includes a terminal device, the third device includes a radio access network device, and the transmission of data includes at least one of the following: uplink transmission from the terminal device to the radio access network device, or downlink transmission from the radio access network device to the terminal device.

[0139] In some exemplary embodiments, the first device includes a first terminal device, the second device includes a second terminal device, the third device includes a radio access network device, and the data transmission includes at least one of the following: a sidelink transmission between the first terminal device and the second terminal device, or a sidelink transmission between the second terminal device and the third terminal device.

[0140] In some example embodiments, the method 500 further includes: receiving assistance information from the first apparatus; determining, based on the assistance information, scheduling information used or to be used for transmission of the scheduled data; and sending the scheduling information to the first apparatus.

[0141] In some example embodiments, the method 500 further comprises receiving, from the first device, a confirmation of the scheduling information.

[0142] In some example embodiments, the first apparatus comprises a terminal device and the second apparatus comprises another terminal device.

[0143] In some example embodiments, the method 500 further includes receiving a request for characteristic information of the data from the first device.

[0144] In some example embodiments, the machine learning model is implemented in at least one of: the first device, the second device, or a third device that schedules data transmission to the second device.

[0145] Figure 6 FIG. 6 is a flowchart of an example method 600 implemented at a third device according to some example embodiments of the present disclosure. Figure 2 The method 600 is described from the perspective of the third device 203.

[0146] At block 610, a third device receives auxiliary information from a first device for scheduling transmission of data to be sent to or from a second device. The data is related to a machine learning model.

[0147] At block 620 , the third device determines scheduling information used or to be used for transmission of the data based on the assistance information.

[0148] In block 630, the third device sends scheduling information to the first device.

[0149] In some example embodiments, the method 600 further comprises receiving, from the first device, a confirmation of the scheduling information.

[0150] In some example embodiments, method 600 further includes receiving, from the first device, an indication of functionality for updating the machine learning model.

[0151] In some exemplary embodiments, the first device includes a core network device or a radio access network device, the second device includes a terminal device, the third device includes a radio access network device, and the transmission of data includes at least one of the following: uplink transmission from the terminal device to the radio access network device, or downlink transmission from the radio access network device to the terminal device.

[0152] In some exemplary embodiments, the first device includes a first terminal device, the second device includes a second terminal device, the third device includes a radio access network device, and the transmission of data includes at least one of the following: a side link transmission between the first terminal device and the second terminal device, or a side link transmission between the second terminal device and the third terminal device.

[0153] In some example embodiments, the machine learning model is implemented in at least one of: the first device, the second device, or the third device.

[0154] In some example embodiments, a first device (eg, Figure 2 and Figure 3 The first device in the method 400 may include a component for performing the corresponding operation of the method 400. The component may be implemented in any suitable form. For example, the component may be implemented in a circuit or a software module. The first device may be implemented as or included in the CN device 111, the terminal device 112, and the terminal device 113.

[0155] In some example embodiments, the first device includes: a component for receiving characteristic information of data to be sent to or from the second device from the second device, wherein the data is related to the machine learning model; and a component for determining auxiliary information for scheduling transmission of the data based at least on the characteristic information.

[0156] In some example embodiments, the first apparatus comprises means for sending assistance information to a third apparatus, the third apparatus scheduling data transmission for the first apparatus; and means for receiving scheduling information for or to be used for transmission of data from the third apparatus.

[0157] In some example embodiments, the first device also includes a component for determining whether the scheduling information meets the quality of service requirements for the functions of the machine learning model; and a component for sending a confirmation of the scheduling information to the third device if it is determined that the scheduling information meets the quality of service requirements.

[0158] In some example embodiments, the first device also includes a component for determining an update to the functionality of the machine learning model based on the scheduling information if it is determined that the scheduling information does not meet the quality of service requirements; and a component for sending an indication for updating the functionality of the machine learning model to at least one of the third device or the second device.

[0159] In some exemplary embodiments, the first device includes a core network device or a radio access network device, the second device includes a terminal device, the third device includes a radio access network device, and the transmission of data includes at least one of the following: uplink transmission from the terminal device to the radio access network device, or downlink transmission from the radio access network device to the terminal device.

[0160] In some exemplary embodiments, the first device includes a first terminal device, the second device includes a second terminal device, the third device includes a radio access network device, and the transmission of data includes at least one of the following: a side link transmission between the first terminal device and the second terminal device, or a side link transmission between the second terminal device and the third terminal device.

[0161] In some example embodiments, the first apparatus comprises means for sending assistance information to the second apparatus; and means for receiving scheduling information for or to be used for transmission of data from the second apparatus.

[0162] In some example embodiments, the first device also includes a component for determining whether the scheduling information meets the quality of service requirements for the functions of the machine learning model; and a component for sending a confirmation of the scheduling information to the second device if it is determined that the scheduling information meets the quality of service requirements.

[0163] In some example embodiments, the first device also includes a component for determining an update to the functionality of the machine learning model based on the scheduling information if it is determined that the scheduling information does not meet the quality of service requirements; and a component for sending an indication to the second device for updating the functionality of the machine learning model.

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

[0165] In some example embodiments, the first device further comprises means for sending a request for characteristic information of the data to the second device.

[0166] In some example embodiments, the machine learning model is implemented in at least one of: the first device, the second device, or a third device that schedules data transmission to the second device.

[0167] In some example embodiments, the first apparatus further comprises means for performing the method 400 or other operations in some example embodiments of the first apparatus 201. 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 first apparatus to perform.

[0168] In some example embodiments, a second device (eg, Figure 2 and Figure 3 The second device in the method 500 may include a device for performing the corresponding operation of the method 500. The device may be implemented in any suitable form. For example, the device may be implemented in a circuit or a software module. The second device may be implemented as or included in the terminal device 120.

[0169] In some example embodiments, the second device includes components for sending, to the first device, characteristic information of data to be sent to or from the second device, wherein the data is associated with a machine learning model, and wherein the characteristic information is used to determine auxiliary information for scheduling the sending of the data.

[0170] In some example embodiments, the second device further comprises means for receiving, from the first device, an indication of functionality for updating the machine learning model.

[0171] In some example embodiments, the second apparatus further includes: means for receiving scheduling information for transmission of the data from the third apparatus; and means for performing the transmission based on the scheduling information.

[0172] In some exemplary embodiments, the first device includes a core network device or a radio access network device, the second device includes a terminal device, the third device includes a radio access network device, and the transmission of data includes at least one of the following: uplink transmission from the terminal device to the radio access network device, or downlink transmission from the radio access network device to the terminal device.

[0173] In some exemplary embodiments, the first device includes a first terminal device, the second device includes a second terminal device, the third device includes a radio access network device, and the data transmission includes at least one of the following: a sidelink transmission between the first terminal device and the second terminal device, or a sidelink transmission between the second terminal device and the third terminal device.

[0174] In some example embodiments, the second apparatus further comprises: means for receiving assistance information from the first apparatus; means for determining, based on the assistance information, scheduling information used or to be used for transmission of the scheduling data; and means for sending the scheduling information to the first apparatus.

[0175] In some example embodiments, the second apparatus further comprises means for receiving an acknowledgement of the scheduling information from the first apparatus.

[0176] In some example embodiments, the first apparatus comprises a terminal device and the second apparatus comprises another terminal device.

[0177] In some example embodiments, the second device further comprises means for receiving a request for characteristic information of the data from the first device.

[0178] In some example embodiments, the machine learning model is implemented in at least one of: the first device, the second device, or a third device that schedules data transmission to the second device.

[0179] In some example embodiments, the second apparatus further comprises means for performing the method 500 or other operations in some example embodiments of the second apparatus 202. 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 second apparatus to perform.

[0180] In some example embodiments, a third device (eg, Figure 2 The third device 203 in the method 600 may include a component for performing the corresponding operation of the method 600. The component may be implemented in any suitable form. For example, the component may be implemented in a circuit or a software module. The third device may be implemented as or included in the RAN device 131 or 132.

[0181] In some example embodiments, the third device includes: a component for receiving auxiliary information from the first device for scheduling the transmission of data to be sent to or to be sent from the second device, wherein the data is related to the machine learning model; a component for determining scheduling information used for or to be used for the transmission of the data based on the auxiliary information; and a component for sending the scheduling information to the first device.

[0182] In some example embodiments, the third apparatus further comprises means for receiving an acknowledgement of the scheduling information from the first apparatus.

[0183] In some example embodiments, the third device further comprises means for receiving, from the first device, an indication of functionality for updating the machine learning model.

[0184] In some exemplary embodiments, the first device includes a core network device or a radio access network device, the second device includes a terminal device, the third device includes a radio access network device, and the transmission of data includes at least one of the following: uplink transmission from the terminal device to the radio access network device, or downlink transmission from the radio access network device to the terminal device.

[0185] In some exemplary embodiments, the first device includes a first terminal device, the second device includes a second terminal device, the third device includes a radio access network device, and the transmission of data includes at least one of the following: a side link transmission between the first terminal device and the second terminal device, or a side link transmission between the second terminal device and the third terminal device.

[0186] In some example embodiments, the machine learning model is implemented in at least one of: the first device, the second device, or the third device.

[0187] In some example embodiments, the third apparatus further comprises means for performing the method 600 or other operations in some example embodiments of the third apparatus 203. In some example embodiments, the means comprises: at least one processor; and at least one memory storing instructions which, when executed by the at least one processor, cause the third apparatus to perform.

[0188] Figure 7 7 is a simplified block diagram of a device 700 suitable for implementing an example embodiment of the present disclosure. The device 700 may be used to implement a communication device, such as the first device 201 or the second device 202, or the third device 203. As shown, the device 700 includes one or more processors 710, one or more memories 720 coupled to the processor 710, and one or more communication modules 740 coupled to the processor 710.

[0189] The communication module 740 is used for two-way communication. The communication module 740 has one or more communication interfaces to support communication with one or more other modules or devices. The communication interface may represent any interface necessary for communication with other network elements. In some example embodiments, the communication module 740 may include at least one antenna.

[0190] As non-limiting examples, processor 710 may be of any type suitable for a local technology network, and may include one or more of the following: a general purpose computer, a special purpose computer, a microprocessor, a digital signal processor (DSP), and a processor based on a multi-core processor architecture. Device 700 may have multiple processors, such as application specific integrated circuit chips that are time slaved to a clock of a synchronized main processor.

[0191] The memory 720 may include one or more non-volatile memories and one or more volatile memories. Examples of non-volatile memories include, but are not limited to, read-only memory (ROM) 724, electrically programmable read-only memory (EPROM), flash memory, hard disk, compact disk (CD), digital video disk (DVD), optical disk, laser disk, and other magnetic storage and / or optical storage. Examples of volatile memories include, but are not limited to, random access memory (RAM) 722 and other volatile memories that will not persist over the duration of a power outage.

[0192] The computer program 730 includes computer executable instructions executed by the associated processor 710. The instructions of the program 730 may include instructions for performing the operations / actions of some example embodiments of the present disclosure. The program 730 may be stored in a memory, such as ROM 724. The processor 710 may perform any suitable actions and processes by loading the program 730 into the RAM 722.

[0193] The exemplary embodiments of the present disclosure may be implemented by means of a program 730, so that the device 700 may execute the Figures 2 to 6 Any process of the present disclosure discussed. The exemplary embodiments of the present disclosure may also be implemented by hardware or by a combination of software and hardware.

[0194] In some example embodiments, the program 730 may be tangibly contained in a computer-readable medium, which may be included in the device 700 (such as in the memory 720) or in other storage devices accessible by the device 700. The device 700 may load the program 730 from the computer-readable medium to the RAM 722 for execution. In some example embodiments, the computer-readable medium may include any type of non-transitory storage medium, such as ROM, EPROM, flash memory, hard disk, CD, DVD, etc. The term "non-transitory" as used herein is a limitation of the medium itself (i.e., tangible, not a signal), not a limitation on data storage persistence (e.g., RAM versus ROM).

[0195] Figure 8 An example of a computer readable medium 800 is shown which may be in the form of a CD, DVD or other optical storage disk. The computer readable medium 800 has a program 730 stored thereon.

[0196] Generally, various embodiments of the present disclosure may be implemented in hardware or dedicated circuits, software, logic, or any combination thereof. Some aspects may be implemented in hardware, and other aspects may be implemented in firmware or software that may be executed by a controller, microprocessor, or other computing device. Although various aspects of the embodiments of the present disclosure are shown and described as block diagrams, flow charts, or using some other graphical representations, it should be understood that, as non-limiting examples, the boxes, devices, systems, techniques, or methods described herein may be implemented in hardware, software, firmware, dedicated circuits or logic, general-purpose hardware or controllers or other computing devices, or some combination thereof.

[0197] 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 computer executable instructions, such as those included in a program module executed in a device on a target physical or virtual processor to perform any method as described above. Typically, a program module includes routines, programs, libraries, objects, classes, components, data structures, etc. that perform specific tasks or implement specific abstract data types. In various embodiments, the functions of the program modules can be combined or split between program modules as needed. Machine executable instructions for program modules can be executed in local or distributed devices. In distributed devices, program modules can be located in both local and remote storage media.

[0198] The program code for executing the method of the present disclosure can be written in any combination of one or more programming languages. The program code can be provided to a processor or controller of a general-purpose computer, a special-purpose computer or other programmable data processing device so that the program code enables the functions / operations specified in the flow chart and / or block diagram to be implemented when executed by the processor or controller. The program code can be executed entirely on the machine, partially on the machine, as an independent software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0199] In the context of the present disclosure, computer program codes or related data may be carried by any suitable carrier to enable a device, apparatus or processor to perform various processes and operations as described above. Examples of carriers include signals, computer readable media, etc.

[0200] The computer readable medium may be a computer readable signal medium or a computer readable storage medium. The computer readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or apparatuses, or any suitable combination of the foregoing. More specific examples of computer readable storage media would include an electrical connection with one or more wires, a portable computer disk, 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 disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0201] In addition, although operations are depicted in a particular order, this should not be understood as requiring such operations to be performed in the particular order shown or in sequence, or performing all of the operations shown to achieve the desired result. In some cases, multitasking and parallel processing may be advantageous. Similarly, although several specific implementation details are included in the above discussion, these should not be interpreted as limitations on the scope of the present disclosure, but as descriptions of features that may be specific to a particular embodiment. Unless explicitly stated, certain features described in the context of a separate embodiment may also be implemented in combination in a single embodiment. On the contrary, unless explicitly stated, the various features described in the context of a single embodiment may also be implemented in multiple embodiments individually or in any suitable sub-combination.

[0202] Although the disclosure has been described in language specific to structural features and / or methodological acts, it should be understood that the 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

1. A second device for communication, comprising: at least one processor; as well as at least one memory storing instructions, which when executed by the at least one processor cause the second device to at least: sending, to the first device, characteristic information of data to be sent to or from the second device, wherein the data is related to the machine learning model, and The characteristic information is used to determine the auxiliary information used to schedule the transmission of the data.

2. The second device of claim 1, wherein the instructions, when executed by the at least one processor, further cause the second device to: An indication of functionality for updating the machine learning model is received from the first device.

3. The second device of claim 1 , wherein the instructions, when executed by the at least one processor, further cause the second device to: receiving scheduling information for the transmission of the data from a third device; and The transmitting is performed based on the scheduling information.

4. The second device according to claim 3, wherein the first device comprises a core network device or a radio access network device, the second device comprises a terminal device, the third device comprises a radio access network device, and The transmission of the data includes at least one of the following: an uplink transmission from the terminal device to the radio access network device, or Downlink transmission from the radio access network equipment to the terminal equipment.

5. The second apparatus according to claim 3, wherein the first apparatus comprises a first terminal device, the second apparatus comprises a second terminal device, the third apparatus comprises a radio access network device, and The data transmission includes at least one of the following: side link transmission between the first terminal device and the second terminal device, or Side link transmission between the second terminal device and the third terminal device.

6. The second device of claim 1, wherein the instructions, when executed by the at least one processor, further cause the second device to: receiving the auxiliary information from the first device; Based on the auxiliary information, determining scheduling information used or to be used for scheduling transmission of the data; and The scheduling information is sent to the first device.

7. The second device of claim 6, wherein the instructions, when executed by the at least one processor, further cause the second device to: An acknowledgement of the scheduling information is received from the first device.

8. The second apparatus according to claim 6, wherein the first apparatus comprises a terminal device and the second apparatus comprises another terminal device.

9. The second device of claim 1, wherein the instructions, when executed by the at least one processor, further cause the second device to: A request for the characteristic information of the data is received from the first device.

10. The second apparatus according to claim 1, wherein the machine learning model is implemented in at least one of the following: the first device, the second device, or A third device schedules a data transmission to the second device.

Images