Prioritization and allocation of tasks

Abstract

Embodiments of the present disclosure relate to apparatuses, methods, and computer readable storage media for prioritization and allocation of tasks. In the method, a first apparatus receives, from a plurality of second apparatuses, a plurality of respective first indications. Each first indication indicates availability information of machine-learning (ML)-related resources of one of the plurality second apparatuses. Then, the first apparatus determines, based on the plurality of respective first indications, a second indication for a second apparatus of the plurality second apparatuses. The second indication is used by the second apparatus for determining at least one task performed by the second apparatus. The at least one task comprises at least one of a ML-related task or a non-ML task. Furthermore, the first apparatus transmits the second indication to the second apparatus. Thereby, the proposed solutions can advantageously enable and achieve efficient utilization of network resources.

Description

PRIORITIZATION AND ALLOCATION OF TASKSFIELDS

[0001] Various example embodiments of the present disclosure generally relate to the field of telecommunication and in particular, to methods, devices, apparatuses and computer readable storage medium for prioritization and allocation of tasks.BACKGROUND

[0002] In some communication systems such as the next cellular systems, artificial intelligence (Al) and / or machine learning (ML) technology is proposed to be used in order to improve the communication performance. An AI / ML model may be applied in the new radio (NR) radio interface to assist model functionalities or communication-related functions, such as, channel state information (CSI) overhead reduction, beam management (BM), positioning, and the like. For example, the AI / ML-based beam management targets spatial and / or time beam prediction for overhead and latency reduction. The functions such as prioritization and allocation of tasks need to be enhanced.SUMMARY

[0003] In a first aspect of the present disclosure, there is provided a first apparatus. The first apparatus comprises at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the first apparatus at least to: receive, from a plurality of second apparatuses, a plurality of respective first indications, each first indication indicating availability information of machine-learning (ML)-related resources of one of the plurality second apparatuses; determine, based on the plurality of respective first indications, a second indication for a second apparatus of the plurality second apparatuses, the second indication being used by the second apparatus for determining at least one task performed by the second apparatus, the at least one task comprising at least one of a ML-related task or a non-ML task; and transmit the second indication to the second apparatus.

[0004] In a second aspect of the present disclosure, there is provided a second apparatus. The second apparatus comprises at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the second apparatusat least to: transmit, to a first apparatus, a first indication indicating availability information of machine-learning (ML)-related resources of the second apparatus; and receive, from the first apparatus, a second indication used by the second apparatus for determining at least one task to be performed by the second apparatus, the at least one task comprising at least one of a ML-related task or a non-ML task, the second indication being determined by the first apparatus based on the first indication and at least one further first indication indicating further availability information of ML-related resources of at least one further second apparatuses.

[0005] In a third aspect of the present disclosure, there is provided a method. The method comprises: receiving, from a plurality of second apparatuses, a plurality of respective first indications, each first indication indicating availability information of machine-learning (ML)-related resources of one of the plurality second apparatuses; determining, based on the plurality of respective first indications, a second indication for a second apparatus of the plurality second apparatuses, the second indication being used by the second apparatus for determining at least one task performed by the second apparatus, the at least one task comprising at least one of a ML-related task or a non-ML task; and transmitting the second indication to the second apparatus.

[0006] In a fourth aspect of the present disclosure, there is provided a method. The method comprises: transmitting, to a first apparatus, a first indication indicating availability information of machine-learning (ML)-related resources of the second apparatus; and receiving, from the first apparatus, a second indication used by the second apparatus for determining at least one task to be performed by the second apparatus, the at least one task comprising at least one of a ML-related task or a non-ML task, the second indication being determined by the first apparatus based on the first indication and at least one further first indication indicating further availability information of ML-related resources of at least one further second apparatuses.

[0007] In a fifth aspect of the present disclosure, there is provided a first apparatus. The first apparatus comprises means for receiving, from a plurality of second apparatuses, a plurality of respective first indications, each first indication indicating availability information of machine-learning (ML)-related resources of one of the plurality second apparatuses; means for determining, based on the plurality of respective first indications, a second indication for a second apparatus of the plurality second apparatuses, the second indication being used by the second apparatus for determining at least one task performedby the second apparatus, the at least one task comprising at least one of a ML-related task or a non-ML task; and means for transmitting the second indication to the second apparatus.

[0008] In a sixth aspect of the present disclosure, there is provided a second apparatus. The second apparatus comprises means for transmitting, to a first apparatus, a first indication indicating availability information of machine-learning (ML)-related resources of the second apparatus; and means for receiving, from the first apparatus, a second indication used by the second apparatus for determining at least one task to be performed by the second apparatus, the at least one task comprising at least one of a ML-related task or a non-ML task, the second indication being determined by the first apparatus based on the first indication and at least one further first indication indicating further availability information of ML-related resources of at least one further second apparatuses.

[0009] In a seventh aspect of the present disclosure, there is provided a computer readable medium. The computer readable medium comprises instructions stored thereon for causing an apparatus to perform at least the method according to the third aspect.

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

[0011] It is to be understood that the Summary section is not intended to identify key or essential features of embodiments of the present disclosure, nor is it intended to be used to limit the scope of the present disclosure. Other features of the present disclosure will become easily comprehensible through the following description.BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Some example embodiments will now be described with reference to the accompanying drawings, where:

[0013] FIG. 1 illustrates an example communication environment in which example embodiments of the present disclosure can be implemented;

[0014] FIG. 2 is a schematic diagram illustrating ML-enabled features according to some example embodiments of the present disclosure;

[0015] FIG. 3 illustrates a signaling chart for prioritization and allocation of tasksaccording to some example embodiments of the present disclosure

[0016] FIG. 4 illustrates a signaling chart for prioritization and allocation of tasks according to some example embodiments of the present disclosure;

[0017] FIG. 5 illustrates a flowchart of a method implemented at a first apparatus according to some example embodiments of the present disclosure;

[0018] FIG. 6 illustrates a flowchart of a method implemented at a second apparatus according to some example embodiments of the present disclosure;

[0019] FIG. 7 illustrates a simplified block diagram of a device that is suitable for implementing example embodiments of the present disclosure; and

[0020] FIG. 8 illustrates a block diagram of an example computer readable medium in accordance with some example embodiments of the present disclosure.

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

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

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

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

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

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

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

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

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

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

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

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

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

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

[0035] The present disclosure relates to AI / ML for air interface. The use cases of focus include: 1) CSI feedback enhancement, e.g., spatial-frequency domain CSI compression using two-sided Al model, time domain CSI prediction using UE sided model; 2) beam management, e.g., spatial-domain downlink beam prediction for Set A of beams based on measurement results of Set B of beams, temporal downlink beam prediction for Set A of beams based on the historic measurement results of Set B of beams; and / or 3) positioning accuracy enhancements, e.g., direct AI / ML positioning (such as, AI / ML model output: UE location, fingerprinting based on channel observation as the input of AI / ML), or AI / ML assisted positioning (such as, AI / ML model output: new measurement and / or enhancement of existing measurement; Line of Sight (LOS) / Non-LOS (NLOS) identification, timing and / or angle of measurement, likelihood of measurement.

[0036] More specifically, the following cases may also be considered:

[0037] Case 1: UE-based positioning with UE-side model, direct AI / ML or AI / ML assisted positioning;

[0038] Case 2a: UE-assisted / LMF-based positioning with UE-side model, AI / ML assisted positioning;

[0039] Case 2b: UE-assisted / LMF-based positioning with LMF-side model, direct AI / ML positioning;

[0040] Case 3a: NG-RAN node assisted positioning with gNB-side model, AI / ML assisted positioning; and

[0041] Case 3b: NG-RAN node assisted positioning with LMF-side model, direct AI / ML positioning.

[0042] One of the objectives of the existing designs is to assess potential specification impact with respect to protocol aspects, e.g., 3GPP radio access network (RAN) 2, considering aspects related to, e.g., capability indication, configuration and control procedures (training / inference), and management of data and AI / ML model, per 3GPP RANI input.

[0043] As used herein, the term “AI / ML model” is defined as a data driven algorithm that applies AI / ML techniques to generate a set of outputs based on a set of inputs. The following aspects, including the definition of components (if needed) and necessity, are studied in Life Cycle Management (LCM) of an AI / ML model: data collection (this also includes associated assistance information, if applicable), model training, functionality / model identification, model transfer, model inference operation, functionality / model selection, activation, deactivation, switching, and fallback operation, decision by the network (either network initiated or UE-initiated and requested to the network), decision by the UE (event-triggered as configured by the network, UE’s decision reported to the network, or UE-autonomous either with UE’s decision reported to the network or without it), functionality / model monitoring, model update (this includes model finetuning, retraining, and re-development via online / offline training), UE capability, or the like.

[0044] The LCM procedure is studied on the basis that an AI / ML model has a model ID with associated information and / or model functionality at least for some AI / ML operations. With respect to UE-side models and UE-part of two-sided models, for AI / ML functionality identification, legacy 3GPP framework of features as a starting point for discussion may be reused. UE indicates supported functionalities / functionality for a given sub-use-case. UE capability reporting is taken as starting point. For AI / ML model identification, models are identified by model ID at the network. UE indicates supported AI / ML models.

[0045] In functionality-based LCM, network indicates activation / deactivation / fallback / switching of AI / ML functionality via 3GPP signalling (e.g., RRC, MAC-CE, DCI). Models need not be identified at the Network, and UE may perform model-level LCM. Whether and how much awareness / interaction NW shouldhave about model-level LCM will be studied. For functionality identification, there may be either one or more than one Functionalities defined within an AI / ML-enabled feature, whereby AI / ML-enabled Feature refers to a Feature where AI / ML may be used.

[0046] For AI / ML functionality identification and functionality-based LCM of UE-side models and / or UE-part of two-sided models, functionality refers to an AI / ML-enabled Feature / FG enabled by configuration(s), where configuration(s) is(are) supported based on conditions indicated by UE capability. Correspondingly, functionality-based LCM operates based on, at least, one configuration of AI / ML-enabled Feature / FG or specific configurations of an AI / ML-enabled Feature / FG.

[0047] After functionality identification, necessity, mechanisms, for UE to report updates on applicable functionality(es) among configured / identified functionality(es), where the applicable functionalities may be a subset of all configured / identified functionalities are studied.

[0048] In model-ID-based LCM, models are identified at the network, and the network, UE, or either may activate / deactivate / select / switch individual AI / ML models via model ID.

[0049] For AI / ML model identification and model-ID-based LCM of UE-side models and / or UE-part of two-sided models, model-ID-based LCM operates based on identified models, where a model may be associated with specific configurations / conditions associated with UE capability of an AI / ML-enabled Feature / FG and additional conditions (e.g., scenarios, sites, and datasets) as determined / identified between UE-side and NW- side.

[0050] From RANI perspective, an AI / ML model identified by a model ID may be logical, and how it maps to physical AI / ML model(s) may be up to implementation. When distinction is necessary for discussion purposes, companies may use the term a logical AI / ML model to refer to a model that is identified and assigned a model ID, and physical AI / ML model(s) to refer to an actual implementation of such a model.

[0051] It is to be studied the mechanisms through which the UE can report updates on applicable UE part and / or UE-side model(s), where the applicable models may be a subset of all identified models, as well as the necessity to perform such updates.

[0052] Examples of ML features and functionalities will be descried here. Every featureis characterized by a list of UE’s conditions. For each feature, a set of specific functionalities can be characterized by a set of specific UE’s condition values. For example, for the feature AIML_direct_DL_CIR_UEside, the functionality 1-01 is characterized by N’t = 64 only, N_port = 2 only, N_TRP = 12 only; and the functionality 1-02 is characterized by N’t = 128 only, N_port = 2,4 only, N_TRP = 1,...,18 only. In addition, for the AIML_assisted_DL_CIR_LMFside, the functionality 2-01 is characterized by intcrmcdiatc_fcaturc= ToA only, N’t = 128 only, N_port = 1 only, N_TRP = 15 only; and the functionality 2-02 is characterized by intermediate_feature= LOS / NLOS indication only, N’t = 128 only, N_port = 2,4 only, N_TRP = 1,...,18 only.

[0053] Thus, based on the specific UE’s conditions and features related to a target UE, the location management function (LMF) or the network may set a specific Functionality X. It should be noted that the numbers used to distinguish functionalities are used only for illustrative purposes in these examples. In an extreme and ideal scenario, a specific feature can be represented by a unique functionality supporting all UE’s conditions listed in the table for one specific UE.

[0054] In the use-cases considered for AI / ML, network entities are involved in the ways descried below. With regard to CSI feedback enhancement, for spatial-frequency domain CSI compression using two-sided Al model, the gNB may use gNB-side AI / ML model for inference, monitor the gNB-side model, and / or (re)train the gNB-side model. In addition, the gNB may also (re)train and monitor an AI / ML model at the UE side (or the UE-side of the two-sided AI / ML model). This may correspond to (re)training a UE-side model and transferring to the UE, or configuring a UE to train its model. For beam management, spatial-domain downlink beam prediction for Set A of beams based on measurement results of Set B of beams, and temporal Downlink beam prediction for Set A of beams based on the historic measurement results of Set B of beams. With regard to positioning accuracy enhancements, for Case 1 UE-based positioning with UE-side model and / or direct AI / ML or AI / ML assisted positioning, LMF may monitor and (re)train the UE side model. For Case 2a UE-assisted / LMF-based positioning with UE-side model and / or AI / ML assisted positioning, LMF may monitor and (re)train the UE side model. In addition, LMF also conducts LMF-based positioning (by using the UE-side model’s output). For Case 2b UE-assisted / LMF-based positioning with LMF-side model and / or direct AI / ML positioning, LMF may use LMF-side model for inference, monitor the LMF- side model, and (re)train the LMF-side model. For Case 3a NG-RAN node assistedpositioning with gNB-side model and / or AI / ML assisted positioning, gNB may use gNB- side AI / ML model for inference, monitor the gNB-side model, and (re)train the gNB-side model. In addition, LMF may also (re)train and monitor the gNB-side model. For Case 3b NG-RAN node assisted positioning with LMF-side model and / or direct AI / ML positioning, the LMF may use LMF-side model for inference, monitor the LMF-side model, and (re)train the LMF-side model.

[0055] As described above, gNB or LMF may need to conduct various AI / ML-related tasks (inference, monitoring, training, etc.), depending on the sidedness (that is where the model runs), which could be UE-side, gNB-side, or LMF-side. Each task, depending on the model and the related functionality, would consume different amounts of ML-related resources (namely processing power, memory, electrical power, etc.) at the gNB-side and LMF-side. Hence, prioritization among different AI / ML tasks at the NW would be necessary in order to reduce (e.g., power, processing) resource consumption for sustainability purposes as well as for making the NW product more attractive to operators.

[0056] For example, training a gNB-side model for CSI compression would consume more CPU resources than another gNB-side model that is used for positioning-related inference due to the relatively frequent update needed for CSI compression compared to positioning; on the other hand, the associated positioning request needs to be satisfied within a much shorter deadline, rendering it more latency critical. When these two tasks conflict, involved gNB(s) would need to make a decision on how to allocate their ML- related resources.

[0057] However, currently, there’s no mechanism to enable efficient coordination and allocation of AI / ML-related tasks (involving different functionalities and models that are mainly related to various ML-enabled features) at the network to enable and achieve efficient utilization of network resources.

[0058] Several solutions are proposed herein to at least address the above-mentioned problem. In the proposed solutions, a first apparatus receives, from a plurality of second apparatuses, a plurality of respective first indications. Each first indication indicates availability information of machine-learning (ML)-related resources of one of the plurality second apparatuses. Then, the first apparatus determines, based on the plurality of respective first indications, a second indication for a second apparatus of the plurality second apparatuses. The second indication is used by the second apparatus fordetermining at least one task performed by the second apparatus. The at least one task comprises at least one of a ML-related task or a non-ML task. Furthermore, the first apparatus transmits the second indication to the second apparatus. The proposed solutions will be described in detail below with reference to FIGS. 1-6. It should be understood that the task may comprise at least one of a ML-related task or a non-ML task, and thus the features, use cases, functionalities, models, and / or methods associated with the task may also be ML-related and / or non-ML related.Example Environment

[0059] FIG. 1 illustrates an example communication environment 100 in which example embodiments of the present disclosure can be implemented. In the communication environment 100, there are a plurality of communication devices, for example, a first apparatus 110 and several second apparatuses 120-1, 120-2, ..., 120-N (which may be individually or collectively referred to as second apparatus 120 hereinafter), where N is an integer. The first apparatus 110 and the second apparatus 120 can communication with each other.

[0060] In the following, for the purpose of illustration, some example embodiments are described with any of the first apparatus 110 and the second apparatus 120 operating as a terminal device (such as, UE or the like), an access network device, a core network device, and / or the like. However, it is to be understood that the above examples are just discussed for purpose of illustrations rather than limitations.

[0061] Communications in the communication environment 100 may be implemented according to any proper communication protocol(s), comprising, but not limited to, cellular communication protocols of the first generation (1G), the second generation (2G), the third generation (3G), the fourth generation (4G), the fifth generation (5G), the sixth generation (6G), and the like, wireless local network communication protocols such as Institute for Electrical and Electronics Engineers (IEEE) 802.11 and the like, and / or any other protocols currently known or to be developed in the future. Moreover, the communication may utilize any proper wireless communication technology, comprising but not limited to: Code Division Multiple Access (CDMA), Frequency Division Multiple Access (FDMA), Time Division Multiple Access (TDMA), Frequency Division Duplex (FDD), Time Division Duplex (TDD), Multiple-Input Multiple -Output (MIMO), Orthogonal Frequency Division Multiple (OFDM), Discrete Fourier Transform spreadOFDM (DFT-s-OFDM) and / or any other technologies currently known or to be developed in the future.

[0062] FIG. 2 is a schematic diagram 200 illustrating ML-enabled features according to some example embodiments of the present disclosure. FIG. 2 provides an interpretation of how ML features, functionalities, and models are linked to each other. For example, the block 210 illustrates a functionality identified by configurations based on conditions #1. For another example, the block 220 illustrates a functionality identified by configurations based on conditions #2. For a further example, the block 230 illustrates a functionality identified by configurations based on conditions #3.

[0063] The potential use of the applicable conditions is shown in FIG. 2. Each functionality may be identified by a subset of applicable conditions, which does not prevent individual applicable conditions to be re-used in different sets and / or different functionalities.Work Principle and Example Signaling for Communication

[0064] FIG. 3 illustrates a signaling chart 300 for prioritization and allocation of tasks according to some example embodiments of the present disclosure. For the purposes of discussion, the signaling chart 300 will be discussed with reference to FIG. 1, for example, by using the first apparatus 110 and the second apparatus 120-1. It is to be understood that, in the example procedure shown in FIG. 3, the second apparatus 120-1 is just an example rather than limitation. The second apparatus 120-1 may be replaced with any of second apparatuses 120-2, ..., 120-N.

[0065] In the signaling chart 300, the second apparatus 120-1 transmit (310), to the first apparatus 110, a first indication indicating availability information of machine-learning (ML)-related resources of the second apparatus. The first apparatus 110 receives (315), from a plurality of second apparatuses, a plurality of respective first indications. Furthermore, the first apparatus 110 determines (320), based on the plurality of respective first indications, a second indication for a second apparatus 120-1 of the plurality second apparatuses. The at least one task comprises at least one of a ML-related task or a non- ML task. The first apparatus 110 transmits (325) the second indication to the second apparatus. The second apparatus 120-1 receives (330) the second indication from the first apparatus 110. The second indication is used by the second apparatus 120-1 for determining at least one task to be performed by the second apparatus.

[0066] In some example embodiments, the availability information of ML-related resources may be associated with a processing capability of the second apparatus, available memory information of the second apparatus, electrical power status of the second apparatus, an input / output condition of the second apparatus, or a communication status of the second apparatus. In addition, the availability information of ML-related resources may be associated with a historical time instance, a present time instance, a future time instance, and / or the like. This will be described in detail below with respect to FIG. 4. It should be understood that the above examples are described merely for purpose of description. The scope of the present disclosure is not limited in this respect.

[0067] In some example embodiments, the availability information of ML-related resources may indicate a flag (e.g., a binary flag or the like), which indicates that the availability information is a dynamically variant information.

[0068] In some example embodiments, the availability information of ML-related resources may indicate area information associated with the availability information. For example, the area information may indicate geographical area, physical area, logical area (e.g., some zones), and / or service area, such as, cell / LMF service area. Additionally or alternatively, the availability information of ML-related resources may indicate a period associated with the availability information. In some further example embodiments, the availability information of ML-related resources may indicate a resource availability pattern associated with the availability information. By way of example, the resource availability pattern may consist of a plurality of time windows.

[0069] In some example embodiments, the availability information of ML-related resources and / or the second indication is associated with a specific use case, a specific feature, a specific functionality, and / or or a specific ML model.

[0070] In some example embodiments, the availability information of ML-related resources may indicate an amount or ratio of ML-related resources dedicated for at least one task associated with a specific use case, a specific feature, a specific functionality, or a specific ML model.

[0071] In some example embodiments, the availability information of ML-related resources may indicate a value of a combined metric calculated based on a function of different availability metrics.

[0072] In some example embodiments, the second indication may indicate a preference order of at least one of the following: a list of use cases, a list of features, a list of functionalities, a list of ML models, or a list of methods. It should be understood that the task may comprise at least one of a ML-related task or a non-ML task, and thus the features, use cases, functionalities, models, and / or methods associated with the task may also be ML-related and / or non-ML related. Additionally or alternatively, the second indication may indicate a priority class corresponding to at least one use case, at least one feature, at least one functionality, at least one ML model, and / or at least one method.

[0073] In some example embodiments, the second indication may indicate an amount of ML-related resource allocated for a specific task, which may be a ML-related task or a non-ML task. Alternatively, the second indication may indicate at least first task to be started by the second apparatus. The first task may be a ML-related task or a non-ML task. In some further example embodiments, the second indication may indicate at least second task to be ceased by the second apparatus. The second task may be a ML-related task or a non-ML task.

[0074] In some example embodiments, the second indication may indicate mappings among at least one range of available resource amount and at least one set of tasks to be performed. In addition, the second indication may further indicate at least one limitation condition of the mappings. The at least one limitation condition may be associated with a specific area, a specific period, a communication quality requirement, and / or the like.

[0075] In some example embodiments, upon receiving (330) the second indication the second apparatus 120-1 may further perform at least one task consistent or inconsistent with the second indication. Moreover, the second apparatus 120-1 may transmit, to the first apparatus 110, a third indication indicating at least one ML-related task and / or information indicating a resource amount allocated to the at least one task. The first apparatus 110 may receive the third indication from the second apparatus.

[0076] In view of the above, in aid of the centralized coordination of AI / ML task allocation and / or prioritization across entities, the available network resources can be utilized more efficiently.

[0077] In view of the above, in aid of the centralized coordination of task allocation and / or prioritization across entities, the available network resources can be utilized more efficiently.Example Embodiments

[0078] Example embodiments of the present disclosure provides a method to prioritize among AI / ML-related tasks across network entities (mainly gNB and LMF) depending on the availability of ML-related resources at the respective entities including processing power, available memory, electrical power, etc.

[0079] For this purpose, a centralized managing entity is proposed. In one example, the centralized managing entity may reside at the core network or radio access network. Additionally, the centralized managing entity may be collocated with a gNB or EMF (or any other RAN node or CN entity). In a further example, the centralized managing entity may be simply implemented as a function of a network entity. In a still further example, the centralized managing entity may also be a part of Operations, Administration and Maintenance (0AM) of network.

[0080] In some embodiments, the managing entity may collect resource availability information from the network entities involved in AI / ML procedures (e.g., different gNBs and LMFs), which might relate to past, current or (expected / predicted) future availability of their ML-related resources. Additionally, involved entities may also report past, current or (expected / predicted) future AI / ML related tasks they are required to perform, such as using an AI / ML model for inference, training a model, etc.

[0081] Given such collected information, the managing entity may determine a policy or configuration that prioritizes different AI / ML-related tasks across the involved entities, as well as offloads or distributes any tasks between the entities. For example, the managing entity swaps the allocation of training a model from one gNB (or the central unit (CU) part of the gNB) to another gNB CU that has more resources available, depending on the resource availability at the entities.

[0082] Entities involved in AI / ML-related tasks then follow the policy / configuration provided to them by the managing entity. With such centralized coordination of AI / ML task allocation and prioritization across entities, available network resources can be utilized more efficiently.

[0083] FIG. 4 illustrates a signaling chart 400 for prioritization and allocation of tasks according to some example embodiments of the present disclosure. In FIG. 4, the managing entity 404 may be an example implementation of the first apparatus 110 in FIG.1. Moreover, each of the network entities 402-1 and 402-2 (which may be individually or collectively referred to as network entity 402 hereinafter) may comprise gNB or LMF, and may be an example implementation of the second apparatus 120 in FIG. 1. It should be understood that although two network entities are shown in FIG. 4, the number of network entities may be any other suitable value. The scope of the present disclosure is not limited in this respect.

[0084] At 410 and 412, the network entities 402-1 and 402-2 indicate ML-related resource availability and responsible AI / ML tasks to the managing entity 404. In other words, each network entity 402 that is involved in a ML-related task indicates to the managing entity 404 its availability information of ML-related resources.

[0085] In some example embodiments, the availability information of ML-related resources may be related to past, present and / or (predicted / expected) future availability of the processing power, e.g., Graphics Processing Unit (GPU), Central Processing Unit (CPU), or Tensor Processing Unit (TPU). Additionally or alternatively, the availability information of ML-related resources may be related to past, present and / or (predicted / expected) future availability of the available memory, e.g., Random Access Memory (RAM). In some further example embodiments, the availability information of ML-related resources may be related to past, present and / or (predicted / expected) future availability of the electrical power, such as, battery status including relevant power level with respect to a threshold level (e.g., the CPU throttling level).

[0086] In some example embodiments, the availability information of ML-related resources may be related to past, present and / or (predicted / expected) future availability of the device / server input / output conditions, e.g., sensor blocked, sensor data missing, etc. By way of example rather than limitation, this applies to the case where sensor-obtained data are used as labels for training Radio Frequency (RF)-based AI / ML positioning methods. Additionally or alternatively, the availability information of ML-related resources may be related to past, present and / or (predicted / expected) future availability of the device / server connection status and / or quality, e.g., Uplink (UL) / Downlink (DL) throughput, link quality, etc.

[0087] In some example embodiments, the resource availability may be indicated with a binary flag as “dynamic”, i.e., the resources used at the entity are not constant but are rather allocated in a dynamic fashion. In some alternative embodiments, the networkentity 402 may indicate resource availability by means of expected time and space (e.g., certain cell area) where the AI / ML-related resources for positioning are available or at certain value (min / max level or range). In one example, the network entity 402 may indicate the geographical bounds (e.g., location of UEs it is serving) for which it expects to deploy AI / ML computational resources above a given threshold. In another example, the network entity 402 may indicate the time windows that are expected to be associated with computational resources above or below a given threshold value. In such a case, the network entity 402 may be aware that during some time of the day there can be more objects in an area than other times of the day (e.g., a busy downtown area which is not busy during afterhours) causing NLOS, which further causes higher CPU demand for using positioning-related ML models. As such, the network entity 402 may share the resource availability pattern associated with prior computations towards AI / ML-based positioning.

[0088] In some further example embodiments, the network entity 402 may indicate the AI / ML-related resources currently allocated for a certain use case, e.g., positioning as a percentage of the total available AI / ML-related resources (e.g., as “20% of my AI / ML processing power is currently dedicated for positioning-related AI / ML tasks”). Alternatively, the network entity 402 may indicate the percentage of the AI / ML-related resources expected to be deployed for a ML use case / feature / functionality / model as a function of a future time window. In some still further example embodiments, the network entity 402 may indicate the percentage of the AI / ML-related resources expected to be deployed for a ML use case / feature / functionality / model as a function of a geographical (or geofenced) area.

[0089] In some alternative embodiments, the network entity 402 may indicate a value of a combined metric, which is calculated as a function of different ML-resource availability metrics, e.g., as a weighted sum of percentages of available CPU power and RAM.

[0090] At 414, the managing entity 404 may determine policy or configuration to prioritize and / or allocate AI / ML tasks across the network entities 402. At 416 and 418, the managing entity 404 may provide the determined policy or configuration for task prioritization or allocation to the respective network entity 402. For example, the policy or configuration may be associated with the ML use case / feature / functionality / model and (available) resources at each network entity 402.

[0091] In some example embodiments, the policy or configuration may indicate a list of ML use case / feature / functionality / model prioritized in the order of preference. That is, which AI / ML-related task to run first in case the ML-related resources are limited. In some further example embodiments, different classes of priorities may be created where each class may consist of multiple ML use cases / features / functionalities / models with the same priority. In this case, network can flexibly select a ML use case / feature / functionality / model within the same priority class by considering additional criteria such as expected maximum Quality of Service (QoS) performance. In some example embodiments, the policy or configuration may limit the number of resources that should be allocated to a specific ML task including different use case / feature / functionality / model.

[0092] In some example embodiments, the policy or configuration may indicate cease or start of a certain ML-related task. Alternatively, the policy or configuration may be a mapping table with the following characteristics. In such a case, the input may comprise a range (e.g., by means of min and max values) of the AI / ML-related computational resources the entity intends to deploy for positioning. Moreover, the output may comprise ML task for a use case / feature / functionality / model as function of the above input. Examples include: making inference for one-step positioning (e.g., a fingerprinting positioning method) versus two-step positioning (e.g., an AI / ML-related method for Line of Sight (LOS) identification followed by a conventional positioning method such as DL / UL-TDoA); iterative refinement of the location estimate by means of hybrid positioning methods (e.g., applying multiple positioning methods towards converging to a high accuracy estimate), or by means of positioning reference unit (PRU)-related correction data.

[0093] In some example embodiments, the network entity 402 may be configured such that in the event the available resources are below some (lower) threshold, the network entity 402 abandons the use case / feature / functionality / model and reverts to a non-AI / ML method (e.g., for positioning use case, using DL-TDoA, UL-TDoA, DL-AoD, etc. via conventional estimation techniques).

[0094] In some example embodiments, the managing entity 404 may indicate a set of conditions that need to be fulfilled for the mapping table to apply. By way of example rather than limitation, such conditions may include level of Signal to Interference plus Noise Ratio (SINR) associated with use-case related measurements (e.g., over PRS / SRSfor positioning, CSI-RS, etc.); or time windows that the mapping table is valid; or area information that the mapping table is valid.

[0095] In some example embodiments, the LMF may assign a list of AI / ML positioning methods which is prioritized in the order of preference. That is, which AI / ML method to use first to calculate the position in case the computational resources are limited.

[0096] At 420, the network entity 402 may run, update, or switch the AI / ML-related task (as well as fallback to non AI / ML methods) as indicated by the policy or configuration.

[0097] In some additional embodiments, the network entity 402 may need to perform further actions as an outcome of the policy or configuration, which are not indicated or different than what is indicated in the policy or configuration. For example, a gNB may select / switch a model to realize a functionality indicated in the policy or configuration, or it may not be able to realize the ML-related task indicated in the policy or configuration. In these cases, at 422 and / or 424, the network entity 402 may indicate to the managing entity 404 about the actions it has taken.

[0098] In view of the above, in aid of the centralized coordination of task allocation and / or prioritization across entities, the available network resources can be utilized more efficiently.Example Methods

[0099] FIG. 5 shows a flowchart of an example method 500 implemented at a first apparatus in accordance with some example embodiments of the present disclosure. For the purpose of discussion, the method 500 will be described from the perspective of the first apparatus 110 in FIG. 1.

[0100] At block 510, the first apparatus 110 receives, from a plurality of second apparatuses, a plurality of respective first indications, each first indication indicating availability information of machine-learning (ML)-related resources of one of the plurality second apparatuses.

[0101] At block 520, the first apparatus 110 determines, based on the plurality of respective first indications, a second indication for a second apparatus of the plurality second apparatuses. The second indication is used by the second apparatus for determining at least one task performed by the second apparatus. The at least one taskcomprises at least one of a ML-related task or a non-ML task.

[0102] At block 530, the first apparatus 110 transmits the second indication to the second apparatus.

[0103] In some example embodiments, the availability information of ML-related resources is associated with at least one of the following: a processing capability of the second apparatus, available memory information of the second apparatus, electrical power status of the second apparatus, an input / output condition of the second apparatus, or a communication status of the second apparatus.

[0104] In some example embodiments, the availability information of ML-related resources is associated with at least one of the following: a historical time instance, a present time instance, or a future time instance.

[0105] In some example embodiments, the availability information of ML-related resources indicates a flag indicating that the availability information is a dynamically variant information.

[0106] In some example embodiments, the availability information of ML-related resources indicates: area information associated with the availability information, a period associated with the availability information, or a resource availability pattern associated with the availability information, the resource availability pattern consisting of a plurality of time windows.

[0107] In some example embodiments, the availability information of ML-related resources and / or the second indication is associated with at least one of the following: a specific use case, a specific feature, a specific functionality, or a specific ML model.

[0108] In some example embodiments, the availability information of ML-related resources indicates an amount or ratio of ML-related resources dedicated for at least one task associated with one of the following: a specific use case, a specific feature, a specific functionality, or a specific ML model.

[0109] In some example embodiments, the availability information of ML-related resources indicates a value of a combined metric calculated based on a function of different availability metrics.

[0110] In some example embodiments, the second indication indicates a preferenceorder of at least one of the following: a list of use cases, a list of features, a list of functionalities, a list of ML models, or a list of methods.

[0111] In some example embodiments, the second indication indicates a priority class corresponding to at least one of the following: at least one use case, at least one feature, at least one functionality, at least one ML model, or at least one method.

[0112] In some example embodiments, the second indication indicates: an amount of ML-related resource allocated for a specific task being a ML-related task or a non-ML task, at least first task to be started by the second apparatus, the first task being a ML- related task or a non-ML task, or at least second task to be ceased by the second apparatus, the second task being a ML-related task or a non-ML task.

[0113] In some example embodiments, the second indication indicates: mappings among at least one range of available resource amount and at least one set of tasks to be performed.

[0114] In some example embodiments, the second indication further indicates at least one limitation condition of the mappings, the at least one limitation condition being associated with at least one of the following: a specific area, a specific period, or a communication quality requirement.

[0115] In some example embodiments, the method 500 further comprises: receiving, from the second apparatus, a third indication indicating at least one of the following: at least one task performed by second apparatus upon receiving the second indication, the at least one task being consistent or inconsistent with the second indication, or information indicating a resource amount allocated to the at least one task.

[0116] In some example embodiments, any of the first and second apparatuses comprises at least one of the following: an access network node, or a core network node.

[0117] FIG. 6 shows a flowchart of an example method 600 implemented at a second apparatus in accordance with some example embodiments of the present disclosure. For the purpose of discussion, the method 600 will be described from the perspective of the second apparatus 120 in FIG. 1.

[0118] At block 610, the second apparatus 120 transmits, to a first apparatus, a first indication indicating availability information of machine-learning (ML)-related resources of the second apparatus 120.

[0119] At block 620, the second apparatus 120 receives, from the first apparatus, a second indication used by the second apparatus 120 for determining at least one task to be performed by the second apparatus 120, the at least one task comprising at least one of a ML-related task or a non-ML task, the second indication being determined by the first apparatus based on the first indication and at least one further first indication indicating further availability information of ML-related resources of at least one further second apparatus 120es.

[0120] In some example embodiments, the availability information of ML-related resources is associated with at least one of the following: a processing capability of the second apparatus 120, available memory information of the second apparatus 120, electrical power status of the second apparatus 120, an input / output condition of the second apparatus 120, or a communication status of the second apparatus 120.

[0121] In some example embodiments, the availability information of ML-related resources is associated with at least one of the following: a historical time instance, a present time instance, or a future time instance.

[0122] In some example embodiments, the availability information of ML-related resources indicates a flag indicating that the availability information is a dynamically variant information.

[0123] In some example embodiments, the availability information of ML-related resources indicates: area information associated with the availability information, a period associated with the availability information, or a resource availability pattern associated with the availability information, the resource availability pattern consisting of a plurality of time windows.

[0124] In some example embodiments, the availability information of ML-related resources and / or the second indication is associated with at least one of the following: a specific use case, a specific feature, a specific functionality, or a specific ML model.

[0125] In some example embodiments, the availability information of ML-related resources indicates an amount or ratio of ML-related resources dedicated for at least one task associated with one of the following: a specific use case, a specific feature, a specific functionality, or a specific ML model.

[0126] In some example embodiments, the availability information of ML-relatedresources indicates a value of a combined metric calculated based on a function of different availability metrics.

[0127] In some example embodiments, the second indication indicates a preference order of at least one of the following: a list of use cases, a list of features, a list of functionalities, a list of ML models, or a list of methods.

[0128] In some example embodiments, the second indication indicates a priority class corresponding to at least one of the following: at least one use case, at least one feature, at least one functionality, at least one ML model, at least one method.

[0129] In some example embodiments, the second indication indicates: an amount of ML-related resource allocated for a specific task being a ML-related task or a non-ML task, at least first task to be started by the second apparatus 120, the first task being a ML- related task or a non-ML task, or at least second task to be ceased by the second apparatus 120, the second task being a ML-related task or a non-ML task.

[0130] In some example embodiments, the second indication indicates: mappings among at least one range of available resource amount and at least one set of tasks to be performed.

[0131] In some example embodiments, the second indication further indicates at least one limitation condition of the mappings, the at least one limitation condition being associated with at least one of the following: a specific area, a specific period, or a communication quality requirement.

[0132] In some example embodiments, the method 600 further comprises: upon receiving the second indication, performing at least one task being consistent or inconsistent with the second indication; and transmitting, to the first apparatus, a third indication indicating at least one of the following: at least one ML-related task, or information indicating a resource amount allocated to the at least one task.

[0133] In some example embodiments, any of the first and second apparatuses comprises at least one of the following: an access network node, or a core network node.Example Apparatus, Device and Medium

[0134] In some example embodiments, a first apparatus capable of performing any of the method 500 (for example, the first apparatus 110 in FIG. 1) may comprise means forperforming the respective operations of the method 500. The means may be implemented in any suitable form. For example, the means may be implemented in a circuitry or software module. The first apparatus may be implemented as or included in the first apparatus 110 in FIG. 1.

[0135] In some example embodiments, the first apparatus comprises means for receiving, from a plurality of second apparatuses, a plurality of respective first indications, each first indication indicating availability information of machine-learning (ML)-related resources of one of the plurality second apparatuses; means for determining, based on the plurality of respective first indications, a second indication for a second apparatus of the plurality second apparatuses, the second indication being used by the second apparatus for determining at least one task performed by the second apparatus, the at least one task comprising at least one of a ML-related task or a non-ML task; and means for transmitting the second indication to the second apparatus.

[0136] In some example embodiments, the availability information of ML-related resources is associated with at least one of the following: a processing capability of the second apparatus, available memory information of the second apparatus, electrical power status of the second apparatus, an input / output condition of the second apparatus, or a communication status of the second apparatus.

[0137] In some example embodiments, the availability information of ML-related resources is associated with at least one of the following: a historical time instance, a present time instance, or a future time instance.

[0138] In some example embodiments, the availability information of ML-related resources indicates a flag indicating that the availability information is a dynamically variant information.

[0139] In some example embodiments, the availability information of ML-related resources indicates: area information associated with the availability information, a period associated with the availability information, or a resource availability pattern associated with the availability information, the resource availability pattern consisting of a plurality of time windows.

[0140] In some example embodiments, the availability information of ML-related resources and / or the second indication is associated with at least one of the following: aspecific use case, a specific feature, a specific functionality, or a specific ML model.

[0141] In some example embodiments, the availability information of ML-related resources indicates an amount or ratio of ML-related resources dedicated for at least one task associated with one of the following: a specific use case, a specific feature, a specific functionality, or a specific ML model.

[0142] In some example embodiments, the availability information of ML-related resources indicates a value of a combined metric calculated based on a function of different availability metrics.

[0143] In some example embodiments, the second indication indicates a preference order of at least one of the following: a list of use cases, a list of features, a list of functionalities, a list of ML models, or a list of methods.

[0144] In some example embodiments, the second indication indicates a priority class corresponding to at least one of the following: at least one use case, at least one feature, at least one functionality, at least one ML model, or at least one method.

[0145] In some example embodiments, the second indication indicates: an amount of ML-related resource allocated for a specific task being a ML-related task or a non-ML task, at least first task to be started by the second apparatus, the first task being a ML- related task or a non-ML task, or at least second task to be ceased by the second apparatus, the second task being a ML-related task or a non-ML task.

[0146] In some example embodiments, the second indication indicates: mappings among at least one range of available resource amount and at least one set of tasks to be performed.

[0147] In some example embodiments, the second indication further indicates at least one limitation condition of the mappings, the at least one limitation condition being associated with at least one of the following: a specific area, a specific period, or a communication quality requirement.

[0148] In some example embodiments, the first apparatus further comprises: means for receiving, from the second apparatus, a third indication indicating at least one of the following: at least one task performed by second apparatus upon receiving the second indication, the at least one task being consistent or inconsistent with the second indication, or information indicating a resource amount allocated to the at least one task.

[0149] In some example embodiments, any of the first and second apparatuses comprises at least one of the following: an access network node, or a core network node.

[0150] In some example embodiments, the first apparatus further comprises means for performing other operations in some example embodiments of the method 500 or the first apparatus 110. In some example embodiments, the means comprises at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the performance of the first apparatus.

[0151] In some example embodiments, a second apparatus capable of performing any of the method 600 (for example, the second apparatus 120 in FIG. 1) may comprise means for performing the respective operations of the method 600. The means may be implemented in any suitable form. For example, the means may be implemented in a circuitry or software module. The second apparatus may be implemented as or included in the second apparatus 120 in FIG. 1.

[0152] In some example embodiments, the second apparatus comprises means for transmitting, to a first apparatus, a first indication indicating availability information of machine-learning (ML)-related resources of the second apparatus; and means for receiving, from the first apparatus, a second indication used by the second apparatus for determining at least one task to be performed by the second apparatus, the at least one task comprising at least one of a ML-related task or a non-ML task, the second indication being determined by the first apparatus based on the first indication and at least one further first indication indicating further availability information of ML-related resources of at least one further second apparatuses.

[0153] In some example embodiments, the availability information of ML-related resources is associated with at least one of the following: a processing capability of the second apparatus, available memory information of the second apparatus, electrical power status of the second apparatus, an input / output condition of the second apparatus, or a communication status of the second apparatus.

[0154] In some example embodiments, the availability information of ML-related resources is associated with at least one of the following: a historical time instance, a present time instance, or a future time instance.

[0155] In some example embodiments, the availability information of ML-relatedresources indicates a flag indicating that the availability information is a dynamically variant information.

[0156] In some example embodiments, the availability information of ML-related resources indicates: area information associated with the availability information, a period associated with the availability information, or a resource availability pattern associated with the availability information, the resource availability pattern consisting of a plurality of time windows.

[0157] In some example embodiments, the availability information of ML-related resources and / or the second indication is associated with at least one of the following: a specific use case, a specific feature, a specific functionality, or a specific ML model.

[0158] In some example embodiments, the availability information of ML-related resources indicates an amount or ratio of ML-related resources dedicated for at least one task associated with one of the following: a specific use case, a specific feature, a specific functionality, or a specific ML model.

[0159] In some example embodiments, the availability information of ML-related resources indicates a value of a combined metric calculated based on a function of different availability metrics.

[0160] In some example embodiments, the second indication indicates a preference order of at least one of the following: a list of use cases, a list of features, a list of functionalities, a list of ML models, or a list of methods.

[0161] In some example embodiments, the second indication indicates a priority class corresponding to at least one of the following: at least one use case, at least one feature, at least one functionality, at least one ML model, at least one method.

[0162] In some example embodiments, the second indication indicates: an amount of ML-related resource allocated for a specific task being a ML-related task or a non-ML task, at least first task to be started by the second apparatus, the first task being a ML- related task or a non-ML task, or at least second task to be ceased by the second apparatus, the second task being a ML-related task or a non-ML task.

[0163] In some example embodiments, the second indication indicates: mappings among at least one range of available resource amount and at least one set of tasks to be performed.

[0164] In some example embodiments, the second indication further indicates at least one limitation condition of the mappings, the at least one limitation condition being associated with at least one of the following: a specific area, a specific period, or a communication quality requirement.

[0165] In some example embodiments, the second apparatus further comprises: means for upon receiving the second indication, performing at least one task being consistent or inconsistent with the second indication; and means for transmitting, to the first apparatus, a third indication indicating at least one of the following: at least one ML-related task, or information indicating a resource amount allocated to the at least one task.

[0166] In some example embodiments, any of the first and second apparatuses comprises at least one of the following: an access network node, or a core network node.

[0167] In some example embodiments, the second apparatus further comprises means for performing other operations in some example embodiments of the method 600 or the second apparatus 120. In some example embodiments, the means comprises at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the performance of the second apparatus.

[0168] FIG. 7 is a simplified block diagram of a device 700 that is suitable for implementing example embodiments of the present disclosure. The device 700 may be provided to implement a communication device, for example, the first apparatus 110 or the second apparatus 120 as shown in FIG. 1. As shown, the device 700 includes one or more processors 710, one or more memories 720 coupled to the processor 710, and one or more communication modules 740 coupled to the processor 710.

[0169] The communication module 740 is for bidirectional communications. The communication module 740 has one or more communication interfaces to facilitate communication with one or more other modules or devices. The communication interfaces may represent any interface that is necessary for communication with other network elements. In some example embodiments, the communication module 740 may include at least one antenna.

[0170] The processor 710 may be of any type suitable to the local technical network and may include one or more of the following: general purpose computers, special purpose computers, microprocessors, digital signal processors (DSPs) and processors based onmulticore processor architecture, as non-limiting examples. The device 700 may have multiple processors, such as an application specific integrated circuit chip that is slaved in time to a clock which synchronizes the main processor.

[0171] The memory 720 may include one or more non-volatile memories and one or more volatile memories. Examples of the non-volatile memories include, but are not limited to, a Read Only Memory (ROM) 724, an electrically programmable read only memory (EPROM), a flash memory, a hard disk, a compact disc (CD), a digital video disk (DVD), an optical disk, a laser disk, and other magnetic storage and / or optical storage. Examples of the volatile memories include, but are not limited to, a random access memory (RAM) 722 and other volatile memories that will not last in the power-down duration.

[0172] A computer program 730 includes computer executable instructions that are executed by the associated processor 710. The instructions of the program 730 may include instructions for performing operations / acts of some example embodiments of the present disclosure. The program 730 may be stored in the memory, e.g., the ROM 724. The processor 710 may perform any suitable actions and processing by loading the program 730 into the RAM 722.

[0173] The example embodiments of the present disclosure may be implemented by means of the program 730 so that the device 700 may perform any process of the disclosure as discussed with reference to FIG. 1 to FIG. 6. The example embodiments of the present disclosure may also be implemented by hardware or by a combination of software and hardware.

[0174] In some example embodiments, the program 730 may be tangibly contained in a computer readable medium which may be included in the device 700 (such as in the memory 720) or other storage devices that are accessible by the device 700. The device 700 may load the program 730 from the computer readable medium to the RAM 722 for execution. In some example embodiments, the computer readable medium may include any types of non-transitory storage medium, such as ROM, EPROM, a flash memory, a hard disk, CD, DVD, and the like. The term “non-transitory,” as used herein, is a limitation of the medium itself (i.e., tangible, not a signal) as opposed to a limitation on data storage persistency (e.g., RAM vs. ROM).

[0175] FIG. 8 shows an example of the computer readable medium 800 which may bein form of CD, DVD or other optical storage disk. The computer readable medium 800 has the program 730 stored thereon.

[0176] Generally, various embodiments of the present disclosure may be implemented in hardware or special purpose circuits, software, logic or any combination thereof. Some aspects may be implemented in hardware, and other aspects may be implemented in firmware or software which may be executed by a controller, microprocessor or other computing device. Although various aspects of embodiments of the present disclosure are illustrated and described as block diagrams, flowcharts, or using some other pictorial representations, it is to be understood that the block, apparatus, system, technique or method described herein may be implemented in, as non-limiting examples, hardware, software, firmware, special purpose circuits or logic, general purpose hardware or controller or other computing devices, or some combination thereof.

[0177] Some example embodiments of the present disclosure also provide at least one computer program product tangibly stored on a computer readable medium, such as a non- transitory computer readable medium. The computer program product includes computerexecutable instructions, such as those included in program modules, being executed in a device on a target physical or virtual processor, to carry out any of the methods as described above. Generally, program modules include routines, programs, libraries, objects, classes, components, data structures, or the like that perform particular tasks or implement particular abstract data types. The functionality of the program modules may be combined or split between program modules as desired in various embodiments. Machine-executable instructions for program modules may be executed within a local or distributed device. In a distributed device, program modules may be located in both local and remote storage media.

[0178] Program code for carrying out methods of the present disclosure may be written in any combination of one or more programming languages. The program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program code, when executed by the processor or controller, cause the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may execute entirely on a machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.

[0179] In the context of the present disclosure, the computer program code or related data may be carried by any suitable carrier to enable the device, apparatus or processor to perform various processes and operations as described above. Examples of the carrier include a signal, computer readable medium, and the like.

[0180] The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable medium may include but not limited to an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the computer readable storage medium would include an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0181] Further, although operations are depicted in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed, to achieve desirable results. In certain circumstances, multitasking and parallel processing may be advantageous. Likewise, although several specific implementation details are contained in the above discussions, these should not be construed as limitations on the scope of the present disclosure, but rather as descriptions of features that may be specific to particular embodiments. Unless explicitly stated, certain features that are described in the context of separate embodiments may also be implemented in combination in a single embodiment. Conversely, unless explicitly stated, various features that are described in the context of a single embodiment may also be implemented in a plurality of embodiments separately or in any suitable sub-combination.

[0182] Although the present disclosure has been described in languages specific to structural features and / or methodological acts, it is to be understood that the present disclosure defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the claims.

Claims

WHAT IS CLAIMED IS:

1. A first apparatus comprising: at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the first apparatus at least to: receive, from a plurality of second apparatuses, a plurality of respective first indications, each first indication indicating availability information of machinelearning (ML)-related resources of one of the plurality second apparatuses; determine, based on the plurality of respective first indications, a second indication for a second apparatus of the plurality second apparatuses, the second indication being used by the second apparatus for determining at least one task performed by the second apparatus, the at least one task comprising at least one of a ML-related task or a non-ML task; and transmit the second indication to the second apparatus.

2. The first apparatus of claim 1, wherein the availability information of ML-related resources is associated with at least one of the following: a processing capability of the second apparatus, available memory information of the second apparatus, electrical power status of the second apparatus, an input / output condition of the second apparatus, or a communication status of the second apparatus.

3. The first apparatus of claim 1, wherein the availability information of ML-related resources is associated with at least one of the following: a historical time instance, a present time instance, or a future time instance.

4. The first apparatus of claim 1 , wherein the availability information of ML-related resources indicates a flag indicating that the availability information is a dynamically variant information.

5. The first apparatus of claim 1, wherein the availability information of ML-related resources indicates: area information associated with the availability information, a period associated with the availability information, or a resource availability pattern associated with the availability information, the resource availability pattern consisting of a plurality of time windows.

6. The first apparatus of claim 1, wherein the availability information of ML-related resources and / or the second indication is associated with at least one of the following: a specific use case, a specific feature, a specific functionality, or a specific ML model.

7. The first apparatus of claim 1, wherein the availability information of ML-related resources indicates an amount or ratio of ML-related resources dedicated for at least one task associated with one of the following: a specific use case, a specific feature, a specific functionality, or a specific ML model.

8. The first apparatus of claim 1, wherein the availability information of ML-related resources indicates a value of a combined metric calculated based on a function of different availability metrics.

9. The first apparatus of claim 1, wherein the second indication indicates apreference order of at least one of the following: a list of use cases, a list of features, a list of functionalities, a list of ML models, or a list of methods.

10. The first apparatus of claim 1, wherein the second indication indicates a priority class corresponding to at least one of the following: at least one use case, at least one feature, at least one functionality, at least one ML model, or at least one method.

11. The first apparatus of claim 1, wherein the second indication indicates: an amount of ML-related resource allocated for a specific task being a ML-related task or a non-ML task, at least first task to be started by the second apparatus, the first task being a ML- related task or a non-ML task, or at least second task to be ceased by the second apparatus, the second task being a ML-related task or a non-ML task.

12. The first apparatus of claim 1, wherein the second indication indicates: mappings among at least one range of available resource amount and at least one set of tasks to be performed.

13. The first apparatus of claim 12, wherein the second indication further indicates at least one limitation condition of the mappings, the at least one limitation conditionbeing associated with at least one of the following: a specific area, a specific period, or a communication quality requirement.

14. The first apparatus of claim 1, wherein the first apparatus is further caused to: receive, from the second apparatus, a third indication indicating at least one of the following: at least one task performed by second apparatus upon receiving the second indication, the at least one task being consistent or inconsistent with the second indication, or information indicating a resource amount allocated to the at least one task.

15. The first apparatus of claim 1, wherein any of the first and second apparatuses comprises at least one of the following: an access network node, or a core network node.

16. A second apparatus comprising: at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the second apparatus at least to: transmit, to a first apparatus, a first indication indicating availability information of machine-learning (ML)-related resources of the second apparatus; and receive, from the first apparatus, a second indication used by the second apparatus for determining at least one task to be performed by the second apparatus, the at least one task comprising at least one of a ML-related task or a non-ML task, the second indication being determined by the first apparatus based on the firstindication and at least one further first indication indicating further availability information of ML-related resources of at least one further second apparatuses.

17. The second apparatus of claim 16, wherein the availability information of ML- related resources is associated with at least one of the following: a processing capability of the second apparatus, available memory information of the second apparatus, electrical power status of the second apparatus, an input / output condition of the second apparatus, or a communication status of the second apparatus.

18. The second apparatus of claim 16, wherein the availability information of ML- related resources is associated with at least one of the following: a historical time instance, a present time instance, or a future time instance.

19. The second apparatus of claim 16, wherein the availability information of ML- related resources indicates a flag indicating that the availability information is a dynamically variant information.

20. The second apparatus of claim 16, wherein the availability information of ML- related resources indicates: area information associated with the availability information, a period associated with the availability information, or a resource availability pattern associated with the availability information, the resource availability pattern consisting of a plurality of time windows.

21. The second apparatus of claim 16, wherein the availability information of ML-related resources and / or the second indication is associated with at least one of the following: a specific use case, a specific feature, a specific functionality, or a specific ML model.

22. The second apparatus of claim 16, wherein the availability information of ML- related resources indicates an amount or ratio of ML-related resources dedicated for at least one task associated with one of the following: a specific use case, a specific feature, a specific functionality, or a specific ML model.

23. The second apparatus of claim 16, wherein the availability information of ML- related resources indicates a value of a combined metric calculated based on a function of different availability metrics.

24. The second apparatus of claim 16, wherein the second indication indicates a preference order of at least one of the following: a list of use cases, a list of features, a list of functionalities, a list of ML models, or a list of methods.

25. The second apparatus of claim 16, wherein the second indication indicates a priority class corresponding to at least one of the following: at least one use case, at least one feature, at least one functionality,at least one ML model, at least one method.

26. The second apparatus of claim 16, wherein the second indication indicates: an amount of ML-related resource allocated for a specific task being a ML-related task or a non-ML task, at least first task to be started by the second apparatus, the first task being a ML- related task or a non-ML task, or at least second task to be ceased by the second apparatus, the second task being a ML-related task or a non-ML task.

27. The second apparatus of claim 16, wherein the second indication indicates: mappings among at least one range of available resource amount and at least one set of tasks to be performed.

28. The second apparatus of claim 27, wherein the second indication further indicates at least one limitation condition of the mappings, the at least one limitation condition being associated with at least one of the following: a specific area, a specific period, or a communication quality requirement.

29. The second apparatus of claim 16, wherein the second apparatus is further caused to: upon receiving the second indication, perform at least one task being consistent or inconsistent with the second indication; and transmit, to the first apparatus, a third indication indicating at least one of the following: at least one ML-related task, orinformation indicating a resource amount allocated to the at least one task.

30. The second apparatus of claim 16, wherein any of the first and second apparatuses comprises at least one of the following: an access network node, or a core network node.

31. A method comprising: receiving, from a plurality of second apparatuses, a plurality of respective first indications, each first indication indicating availability information of machine-learning (ML)-related resources of one of the plurality second apparatuses; determining, based on the plurality of respective first indications, a second indication for a second apparatus of the plurality second apparatuses, the second indication being used by the second apparatus for determining at least one task performed by the second apparatus, the at least one task comprising at least one of a ML-related task or a non-ML task; and transmitting the second indication to the second apparatus.

32. A method comprising: transmitting, to a first apparatus, a first indication indicating availability information of machine-learning (ML)-related resources of the second apparatus; and receiving, from the first apparatus, a second indication used by the second apparatus for determining at least one task to be performed by the second apparatus, the at least one task comprising at least one of a ML-related task or a non-ML task, the second indication being determined by the first apparatus based on the first indication and at least one further first indication indicating further availability information of ML-related resources of at least one further second apparatuses.

33. A first apparatus comprising:means for receiving, from a plurality of second apparatuses, a plurality of respective first indications, each first indication indicating availability information of machinelearning (ML)-related resources of one of the plurality second apparatuses; means for determining, based on the plurality of respective first indications, a second indication for a second apparatus of the plurality second apparatuses, the second indication being used by the second apparatus for determining at least one task performed by the second apparatus, the at least one task comprising at least one of a ML-related task or a non-ML task; and means for transmitting the second indication to the second apparatus.

34. A second apparatus comprising: means for transmitting, to a first apparatus, a first indication indicating availability information of machine-learning (ML)-related resources of the second apparatus; and means for receiving, from the first apparatus, a second indication used by the second apparatus for determining at least one task to be performed by the second apparatus, the at least one task comprising at least one of a ML-related task or a non-ML task, the second indication being determined by the first apparatus based on the first indication and at least one further first indication indicating further availability information of ML-related resources of at least one further second apparatuses.

35. A computer readable medium comprising instructions stored thereon for causing an apparatus at least to perform the method of any of claims 31-32.

Images