Apparatus, method, and computer program

Abstract

Data for machine learning is provided wirelessly between two apparatus. A transmission configuration is provided 4005 to the sending apparatus 401, which associates a link quality with a prioritisatio

Description

Field of the disclosure

[0001] The examples described herein generally relate to apparatus, methods, and computer programs, and more particularly (but not exclusively) to apparatus, methods and computer programs for apparatuses. Background

[0002] A communication system can be seen as a facility that enables communication sessions between two or more entities such as communication devices, base stations and / or other nodes by providing carriers between the various entities involved in the communications path.

[0003] The communication system may be a wireless communication system. Examples of wireless systems comprise public land mobile networks (PLMN) operating based on radio standards such as those provided by 3GPP, satellite based communication systems and different wireless local networks, for example wireless local area networks (WLAN). The wireless systems can typically be divided into cells, and are therefore often referred to as cellular systems.

[0004] The communication system and associated devices operate in accordance with a given set of standards or specifications that set out what the various entities associated with the system are permitted to do and how that is to be achieved. Communication protocols and / or parameters that are to be used for the connection are also typically defined. Examples of standards are the so-called 5G standards. Summary

[0005] According to a first aspect, there is provided a first apparatus comprising means for performing: determining that data for use for a machine learning purpose is to be provided over a wireless link between a second apparatus and a third apparatus; determining a transmission configuration for the second apparatus, wherein the transmission configuration associates a link quality of the wireless link with a prioritisation for transmitting the data; and providing the transmission configuration to the second apparatus.

[0006] According to a second aspect, there is provided a second apparatus comprising means for performing: receiving, from a first apparatus, a transmission configuration, wherein the transmission configuration associates a link quality of a wireless link between the second apparatus and a third apparatus with a prioritisation for transmitting the data; determining a link quality of the wireless link; selecting a portion of the data to be transmitted over the wireless link using the transmission configuration; and transmitting the selected portion to the third apparatus over the wireless link.

[0007] According to a third aspect, there is provided a first apparatus comprising: at least one processor; and at least one memory comprising code that, when executed by the at least one processor, causes the first apparatus to perform: determining that data for use for a machine learning purpose is to be provided over a wireless link between a second apparatus and a third apparatus; determining a transmission configuration for the second apparatus, wherein the transmission configuration associates a link quality of the wireless link with a prioritisation for transmitting the data; and providing the transmission configuration to the second apparatus.

[0008] According to a fourth aspect, there is provided a second apparatus comprising: at least one processor; and at least one memory comprising code that, when executed by the at least one processor, causes the second apparatus to perform: receiving, from a first apparatus, a transmission configuration, wherein the transmission configuration associates a link quality of a wireless link between the second apparatus and a third apparatus with a prioritisation for transmitting the data; determining a link quality of the wireless link; selecting a portion of the data to be transmitted over the wireless link using the transmission configuration; and transmitting the selected portion to the third apparatus over the wireless link.

[0009] According to a fifth aspect, there is provided a method for a first apparatus, the method comprising: determining that data for use for a machine learning purpose is to be provided over a wireless link between a second apparatus and a third apparatus; determining a transmission configuration for the second apparatus, wherein the transmission configuration associates a link quality of the wireless link with a prioritisation for transmitting the data; and providing the transmission configuration to the second apparatus.

[0010] According to a sixth aspect, there is provided a method for a second apparatus, the method comprising: receiving, from a first apparatus, a transmission configuration, wherein the transmission configuration associates a link quality of a wireless link between the second apparatus and a third apparatus with a prioritisation for transmitting the data; determining a link quality of the wireless link; selecting a portion of the data to be transmitted over the wireless link using the transmission configuration; and transmitting the selected portion to the third apparatus over the wireless link.

[0011] According to a seventh aspect, there is provided an apparatus comprising: determining circuitry for determining that data for use for a machine learning purpose is to be provided over a wireless link between a second apparatus and a third apparatus; determining circuitry for determining a transmission configuration for the second apparatus, wherein the transmission configuration associates a link quality of the wireless link with a prioritisation for transmitting the data; and providing circuitry for providing the transmission configuration to the second apparatus.

[0012] According to an eighth aspect, there is provided a second apparatus comprising: receiving circuitry for receiving, from a first apparatus, a transmission configuration, wherein the transmission configuration associates a link quality of a wireless link between the second apparatus and a third apparatus with a prioritisation for transmitting the data; determining circuitry for determining a link quality of the wireless link; selecting circuitry for selecting a portion of the data to be transmitted over the wireless link using the transmission configuration; and transmitting circuitry for transmitting the selected portion to the third apparatus over the wireless link.

[0013] According to a ninth aspect, there is provided non-transitory computer readable medium comprising program instructions for causing a first apparatus to perform:: determining that data for use for a machine learning purpose is to be provided over a wireless link between a second apparatus and a third apparatus; determining a transmission configuration for the second apparatus, wherein the transmission configuration associates a link quality of the wireless link with a prioritisation for transmitting the data; and providing the transmission configuration to the second apparatus.

[0014] According to a tenth aspect, there is provided non-transitory computer readable medium comprising program instructions for causing a second apparatus to perform: receiving, from a first apparatus, a transmission configuration, wherein the transmission configuration associates a link quality of a wireless link between the second apparatus and a third apparatus with a prioritisation for transmitting the data; determining a link quality of the wireless link; selecting a portion of the data to be transmitted over the wireless link using the transmission configuration; and transmitting the selected portion to the third apparatus over the wireless link.

[0015] According to an eleventh aspect, there is provided a computer program product stored on a medium that may cause an apparatus to perform any method as described herein.

[0016] According to a twelfth aspect, there is provided an electronic device that may comprise an apparatus as described herein.

[0017] According to a thirteenth third aspect, there is provided a chipset that may comprise an apparatus as described herein. Brief description of FIGs.

[0018] Some examples, will now be described, merely by way of illustration only, with reference to the accompanying drawings in which:

[0019] FIG. 1 shows a schematic representation of a 5G system;

[0020] FIG. 2 shows a schematic representation of a network apparatus;

[0021] FIG. 3 shows a schematic representation of a user equipment;

[0022] FIGs. 4 and 5 illustrate example signalling; and

[0023] FIGs. 6 and 7 illustrate example operations that may be performed by apparatus described herein. Detailed description

[0024] The following describes operations that may be performed in relation to transmissions for data that will be used for artificial intelligence- (AI-) and / or machine learning (ML)-based use cases. Data that is used for Al and / or ML applications is also referred to herein as “AI / ML data”.

[0025] In particular, there is provided a mechanism that evaluates whether a communication link over which such data is to be transmitted is able to transmit data using at least a minimum throughput. When the communication link is unable to support at least the minimum throughput, only some (e.g., not all) of the AI / ML data is transmitted across the link. The un-transmitted AI / ML data may be transmitted at some later time and / or discarded.

[0026] To perform this, the transmitter of the data makes a selection of the AI / ML data to be transmitted. This selection may be performed using a configuration received from a management entity, such as, for exampie, a network function in a 5G core (see below), an access node controller located in a radio access network (RAN), and / or a UE (or some peer apparatus) operating as a server and / or controller of communications of the transmitter of the AI / ML data. The configuration may instruct the transmitter of the AI / ML data how to prioritize AI / ML data for transmission, and when this prioritization is to be applied.

[0027] Further illustration of these techniques is provided below.

[0028] In the following description of examples, certain aspects are explained with reference to devices that are often capable of communication via a wireless cellular system and mobile communication systems serving such mobile communication devices. For brevity and clarity, the following describes such aspects with reference to a 5G wireless communication system. However, it is understood that such aspects are not limited to 5G wireless communication systems, and may, for example, be applied to other wireless communication systems (for example, current 6G proposals, IEEE 802.11, etc.).

[0029] Before describing in detail the examples, certain general principles of a 5G wireless communication system are briefly explained with reference to FIG. 1.

[0030] 3GPP standards defined a service-based architecture in 5G, which is expected to be utilized in 6G and beyond. In a service-based architecture, a modular framework is used in which common applications can be deployed using components from different sources and / or suppliers.

[0031] FIG. 1 shows a schematic representation of a 5G system (5GS) 100. The 5GS may comprise a user equipment (UE) 102 (which may also be referred to as a communication device or a terminal), a 5G access network (AN) (which may be a 5G Radio Access Network (RAN) or any other type of 5G AN such as a Non-3GPP Interworking Function (N3IWF) / a Trusted Non3GPP Gateway Function (TNGF) for Untrusted / Trusted Non-3GPP access or Wireline Access Gateway Function (W-AGF) for Wireline access) 104, a 5G core (5GC) 106, one or more application functions (AF) 108 and one or more data networks (DN) 110.

[0032] The 5G RAN may comprise one or more gNodeB (gNB) distributed unit functions connected to one or more gNodeB (gNB) centralized unit functions. The RAN may comprise one or more access nodes.

[0033] The 5GC 106 may comprise one or more Access and Mobility Management Functions (AMF) 112, one or more Session Management Functions (SMF) 114, one or more authentication server functions (AUSF) 116, one or more unified data management (UDM) functions 118, one or more user plane functions (UPF) 120, one or more unified data repository (UDR) functions 122, one or more network repository functions (NRF) 128, and / or one or more network exposure functions (NEF) 124. The role of an NEF is to provide secure exposure of network services (e.g. voice, data connectivity, charging, subscriber data, and so forth) towards a 3rd party. Although NRF 128 is not depicted with its interfaces, it is understood that this is for clarity reasons and that NRF 128 may have a plurality of interfaces with other network functions.

[0034] The 5GC 106 also comprises a network data analytics function (NWDAF) 126. The NWDAF is responsible for providing network analytics information upon request from one or more network functions or apparatus within the network. Network functions can also subscribe to the NWDAF 126 to receive information therefrom. Accordingly, the NWDAF 126 is also configured to receive and store network information from one or more network functions or apparatus within the network. The data collection by the NWDAF 126 may be performed based on at least one subscription to the events provided by the at least one network function.

[0035] Reference in the below is made to an LMF and an NWDAF in relation to positioning-based operations (e.g., operations relating to determining a location of an apparatus). Therefore, the following provides some additional context in relation to operations relating to positioning-based operations.

[0036] LMF operations in data collection for positioning operations are described in at least 3GPP TS 37.355, TS 38.305, and TS 23.273.

[0037] According to these sections, an LMF can request (and provide to another network function) capabilities of a UE in relation to positioning operations, request (and provide) assistance data (e.g., reference nodes and configuration of positioning signal), and request (and provide) location information. The location information may comprise, for example, location measurement data and / or a location estimate received from the target. The LMF may use the Location Positioning Protocol (LPP) to communicate with a target device (UE), and a New Radio Positioning Protocol a (NRPPa) to communicate with an access node, such as a gNB.

[0038] The LMF decides on the positioning method to be used by an apparatus for determining a location of either that apparatus or of another apparatus. The positioning method may be selected based on, for example, quality of service (QoS) requirements, and / or UE and / or access node capabilities. Measurements for performing the positioning method may be triggered to be made at the UE and / or access node gNB, either via: UE-based positioning (in which a UE calculates the location estimate); and / or UE-assisted / network-based positioning (in which a UE and / or an access node provides measurements data to the IMF, and the LMF calculates the location estimate using those measurement data). The LMF may combine all results and determine a single location estimate. The LMF may also determine an accuracy of the estimate and a velocity of the apparatus whose location is being determined.

[0039] Positioning reference units (PRUs) are UEs with known location. PRUs may support the provision of uplink and / or downlink positioning measurements by providing such measurements to the LMF, which can compare a location obtained using these measurements with a known PRU location to determine correction terms for other nearby UEs.

[0040] NWDAF involvement in data collection for positioning mechanisms is described in 3GPP TS 23.288.

[0041] According to this document, an NWDAF may collect location information for a target UE or a group of target UEs from a location service (LCS) system. The collected location related information may comprise at least one of: a location estimate of the UE; a time stamp of when the location estimate was made; an estimated velocity of the UE (e.g., as in TS 23.032); information about the positioning method used to obtain the location estimate of the UE; an indication of an area event (including when the UE enters, is within or leaves the geographical area associated with the area event); and an indication of motion event when UE moves by more than some predefined straight line distance from a previous location.

[0042] The NWDAF may check user consent using an indicated purpose for the data collection and usage of the data being collected. The NWDAF may check the user consent by retrieving the user consent information from the unified data management (UDM) function. The user consent is subscription information stored in the UDM, and may comprise an indication as to: whether the user authorizes the collection and usage of its data for a particular purpose; and the purpose for which data collection is authorized, e.g. analytics or model training.

[0043] The network may further comprise a management data analytics service (MDAS) producer or MDAS Management Service (MnS) producer. The MDAS MnS producer may provide data analytics in the management plane considering parameters including, for example, load level and / or resource utilization. For example, the MDAS MnS producer for a network function (NF) may collect the NF’s load-related performance data, e.g., resource usage status of the NF. The analysis of the collected data may provide forecast of resource usage information in a predefined future time window. This analysis may also recommend appropriate actions e.g., scaling of resources, admission control, load balancing of traffic, and so forth.

[0044] FIG. 2 shows an example of a control apparatus for a communication system, for example to be coupled to and / or for controlling a station of an access system, such as a RAN node, e.g. a base station, gNB, a central unit of a cloud architecture or a node of a core network such as an Mobility Management Entity (MME) or Serving Gateway (S-GW), a scheduling entity such as a spectrum management entity, or a server or host, for example an apparatus hosting a network repository function (NRF), Network data analytics function (NWDAF), Access and Mobility Management Function (AMF), Session Management Function (SMF), Unified Data Management / Unified Data Repository (UDM / UDR), and so forth. The control apparatus may be integrated with or external to a node or module of a core network or Radio Access Network (RAN). In some examples, base stations comprise a separate control apparatus unit or module. In other examples, the control apparatus can be another network element, such as a radio network controller or a spectrum controller. The control apparatus 200 can be arranged to provide control on communications in the service area of the system. The apparatus 200 comprises at least one memory 201, at least one data processing unit 202, 203 and an input / output interface 204. Via the interface the control apparatus can be coupled to a receiver and a transmitter of the apparatus. The receiver and / or the transmitter may be implemented as a radio front end or a remote radio head. For example, the control apparatus 200 or processor 201 can be configured to execute an appropriate software code to provide the control functions. References to “code” herein are understood to refer to software code, and vice versa.

[0045] The station of the access system may be categorised into two different types: distributed units (DUs), and centralised units (CUs).

[0046] A possible wireless communication device will now be described in more detail with reference to FIG. 3 showing a schematic, partially sectioned view of a communication device 300. Such a communication device is often referred to as user equipment (UE) or terminal. An appropriate mobile communication device may be provided by any device capable of sending and receiving radio signals. Non-limiting examples comprise a mobile station (MS) or mobile device such as a mobile phone or what is referred to as a ’smart phone’, a computer provided with a wireless interface card or other wireless interface facility (e.g., USB dongle), personal data assistant (PDA) or a tablet provided with wireless communication capabilities, or any combinations of these or the like. A mobile communication device may provide, for example, communication of data for carrying communications such as voice, electronic mail (email), text message, multimedia and so on. Users may thus be offered and provided numerous services via their communication devices. Non-limiting examples of these services comprise two-way or multi-way calls, data communication or multimedia services or simply an access to a data communications network system, such as the Internet. Users may also be provided broadcast or multicast data. Nonlimiting examples of the content comprise downloads, television and radio programs, videos, advertisements, various alerts and other information.

[0047] A wireless communication device may be for example a mobile device, that is, a device not fixed to a particular location, or it may be a stationary device. The wireless device may need human interaction for communication, or may not need human interaction for communication. As described herein, the terms UE or “user” are used to refer to any type of wireless communication device.

[0048] The wireless device 300 may receive signals over an air or radio interface 307 via appropriate apparatus for receiving and may transmit signals via appropriate apparatus for transmitting radio signals. In FIG. 3, a transceiver apparatus is designated schematically by block 306. The transceiver apparatus 306 may be provided, for example, by means of a radio part and associated antenna arrangement. The antenna arrangement may be arranged internally or externally to the wireless device.

[0049] A wireless device is typically provided with at least one data processing entity 301, at least one memory 302 and other possible components 303 for use in software code and hardware aided execution of tasks it is designed to perform, including control of access to and communications with access systems and other communication devices. The data processing, storage and other relevant control apparatus can be provided on an appropriate circuit board and / or in chipsets. This feature is denoted by reference 304. The user may control the operation of the wireless device by means of a suitable user interface such as keypad 305, voice commands, touch sensitive screen or pad, combinations thereof or the like. A display 308, a speaker and a microphone can be also provided. Furthermore, a wireless communication device may comprise appropriate connectors (either wired or wireless) to other devices and / or for connecting external accessories, for example hands-free equipment, thereto.

[0050] 3GPP has issued a number of releases (Rei) for defining operating communication protocols related to a communications network. Currently, objectives and work are being set in relation to Releases 18 (Rei. 18) and 19 (Rei. 19).

[0051] As mentioned above, the following relates to data reporting of AI / ML data.

[0052] For clarity and brevity, the following describes techniques in the context of positioning-related AI / ML techniques. However, it is understood that this is merely an example, and that the same principles described below may be applied in respect of any AI / ML-based data collection and / or reporting.

[0053] 3GPP has been considering the use of artificial intelligence and / or machine learning (AI / ML) for assisting with positioning mechanisms for determining a location of an apparatus.

[0054] For example, 3GPP TR 38.843 has considered how AI / ML may be used directly (e.g., determining a UE location using fingerprinting (e.g., mapping) based on channel observation as an input of an AI / ML model) and indirectly (e.g., by assisting determining the value of a parameter that is later used to determine a location of an apparatus, such as, for example, line-of-sight (LOS) or non-line-of-sight (NLOS) determination, angle of measurement, etc.).

[0055] ML-based methods that are currently used for positioning determinations use large amounts of data, mainly fortraining, monitoring, and retraining an ML model. In various types of AI / ML positioning mechanisms, the data used may comprise positioning-related measurements.

[0056] In more detail, training data collection for AI / ML-based positioning information may comprise at least one of the following associated information of training data: Ground truth label for model training, which comprises a report from the label data generation entity: For example, a label data generation entity may label each data sample with actual values to train machine learning models. Stated differently, an ML model may estimate a UE location given a first set of measurements. The ML model may subsequently be trained using data comprising both measurement values and an associated labels of a correct UE location corresponding to those measurement values. a measurement value (corresponding to model input) for model training; report from the measurement data generation entity a quality indicator for and / or associated with ground truth label and / or measurement for model training; report from the label and / or the measurement data generation entity and / or as request from a different (e.g., data collection, etc.) entity. Reference signal configuration(s) used for deriving measurement value Request from data generation entity (e.g., at least one of a UE, PRU, or transmission reception point (TRP)) to the LMF and / or as LMF assistance signalling to the data generation entity. a time stamp for and / or associated with training data for model training; report from data generation entity together with training data and / or as LMF assistance signalling. a separate time stamp for measurement and ground truth label when measurement and ground truth label are generated by different entities. Assistance signalling and procedure to facilitate generating / coilecting training data: Potential determination of the UE / PRU / TRP which can provide the training data Configuration of reference signal (for measurement and / or label) Signalling other than above 2 for data collection, e.g., requested quality of training data

[0057] In addition to training, data collection features for AI / ML training data have been also identified for model monitoring purposes in 3GPP TR 38.843. These include, for example, determining whether a certain type of data is necessary for computing monitoring metric, how an entity can be used to provide the given type of data for calculating monitoring metric: companies requested to report their assumption of the entity (or entities) used to provide the given type of data for calculating monitoring metric for each case, potential signalling to be used for provisioning of the given type of data for calculating an associated monitoring metric, and potential assistance signalling and procedure to facilitate an entity providing data for calculating the monitoring metric. The monitoring metric may comprise a value indicative of the quality of the performance of the ML model. For example, a monitoring metric may comprise a value indicating an accuracy of the ML model’s estimation (e.g., a UE location estimate accuracy).

[0058] Example positioning-related measurements performed may comprise, for example, at least one of: time-based measurements, angle-based measurements, phased-based measurements, or power-based measurements. These measurements may be performed by at least one a UE (e.g., using a downlink positioning reference signal (PRS)) or by an access node (e.g., using any of an uplink sounding reference signal (SRS), an uplink channel impulse response (CIR), or a power delay profile (PDP)).

[0059] Aside from having positioning related measurement data input to the ML model, “ground truth” information associated with them may be used as labels for training supervised ML models. Example “ground truth” (such as UE location information, LOS / NLOS channel type indication, etc.).

[0060] The volume of data that is transferred in a network for ML-related purposes may, in fact, be even larger when it is considered that, besides positioning, other ML-based approaches such as beam management and channel state information feedback are currently being studied in Rel-18 for provision into other AI / ML-based models for helping networks use resources more efficiently. Stated differently, a large amount of data is expected to be transferred between UE and network entities in the upcoming 3GPP releases for addressing the multiple ML-assisted use cases.

[0061] A good link quality between network entities would be useful for transferring such data in large volumes across those network entities. In particular, for network functions in a 5GC that provide data to a UE (e.g., a location management function (LMF), an NWDAF, etc.), the downlink and / or the uplink between a UE and an access node providing access between the UE and the 5GC should usefully support high data rates (e.g., throughput, as well as low latency).

[0062] However, due to mobile and dynamic characteristics of the network, uplink and / or downlink quality might vary significantly across the duration of the connection between the UE and the access node. For example, cell-edge UEs might suffer from low Signal-and-lnterference-to-Noise-Ratio (SINR), hence experience lower throughput as compared to cell-center UEs. Similarly, a UE might be allocated with insufficient resources for its data traffic. Further, when UEs are mobile, handovers of the UE amongst different access nodes would likely impact the communication link quality between the UE and the 5GC.

[0063] Under such conditions, not all of the intended data might be transferred in a timely manner between a UE and the network. This may negatively impact the operation of ML-based mechanisms. To illustrate, in simplest terms, a smaller dataset provided to the UE for training its ML model for positioning might result in worse positioning accuracy than a model trained using a complete set.

[0064] The following aims to address this using at least one of the following described features. These features may relate to adapting data collection for AI / ML purposes between apparatus by taking into account the link conditions on an air interface between a UE and the network.

[0065] For example, when the link conditions between the UE and the network are undesirable, then only a minimum amount of AI / ML data (e.g. labelled data, as discussed above) may be transferred between the UE and the network. Labelled data may be considered more useful than unlabelled data. Therefore, more useful data for the model (such as labelled data) may be signalled when the link conditions are relatively poor. Stated differently, in an example, more useful data for an ML model may be prioritized for transmission over less useful data during poor link conditions.

[0066] The minimum amount of AI / ML data may be determined based on data that is labelled as being a high priority relative to the priorities associated with other labelled AI / ML data. Examples of high priority data are parts of data lacking at the receiver, parts of data with lower redundancy, data providing better generalization performance, etc.

[0067] The minimum amount of AI / ML data may be determined based on an indication of a maximum volume of AI / ML data (or an indication of a maximum volume of traffic) that may be transmitted.

[0068] Stated differently, in undesirable network conditions (e.g., when a value of a quality metric (e.g., signal-to-noise ratio) falls below a predetermined threshold amount), instead of transmitting a total dataset of data requested, only a subset (e.g., less than all) of the total dataset is transmitted. This subset may be selected based on a determination of relative priorities of different data within the dataset.

[0069] In the following, AI / ML data associated with positioning is considered as an example. However, it is understood that the presently described techniques may be applied to any other type of use case for which data is being provided for AI / ML purposes across a channel or link that is subject to variable quality. Further, it is understood that the presently described techniques are not limited to the management functions described herein (e.g., an LMF and / or an NWDAF), but may be performed by any management function involved in AI / ML data collection.

[0070] There is described a management function (e.g., an LMF and / or an NWDAF) that controls data collection, and determines an “AI / ML data collection configuration”. The AI / ML data collection configuration is a mapping between data to be collected between a UE and a network (e.g., certain parts or characteristics of data), and the link quality between an access point to the network and the UE (e.g., in terms of throughput). The link between the network and the UE is currently labelled as a Uu link.

[0071] The AI / ML data collection configuration may be determined as an ordered list of pairs (e.g., as an AI / ML data priority - link quality pair). This may enable higher priority data to be sent first.

[0072] The AI / ML data priority may be dynamic. In such a case, the AI / ML data priority for a specific type of data may be adjusted based on at least one factor. The at least one factor may comprise, for example, data availability, and / or a need for data with certain characteristics at the data collection entity (where the “need” may be determined based on an identified lack (or overabundance) of that data type relative to other data types used by the ML / AI model).

[0073] For the case of uplink data transfer from the UE to the network, the network may provide the UE with an AI / ML data configuration determined by the management function for uplink data traffic. The UE may monitor uplink link quality and / or uplink link availability, and transfer AI / ML data to the network (via an access node) in accordance with (e.g., based on) the mapping and priority comprised in the received AI / ML data configuration.

[0074] With reference to a first positioning example, an LMF may configure a UE to conduct downlink PRS measurements from 24 different Transmission-Reception Points (TRPs) for use in estimating the UE’s location. However, due to a low link quality between a UE and its serving gNB, not all of these measurements can be reported at once. Considering this, the LMF configures UE to report only a subset (e.g., less than all of the measurements requested) of the measurements first. For the present example, this may result in PRS measurements being transmitted upstream by the UE in respect of 8 specific TRPs of the 24 TRPs when the link quality is determined by the UE to be below a threshold, and to PRS measurements being transmitted upstream for all 24 of the TRPs when the link quality is determined by the UE to be above the threshold.

[0075] For the case of downlink transfer of data from network to UE (via at least one access node), the network provides an AI / ML data configuration to the at least one access node. Subsequently, the access node may monitor downlink link quality and / or downlink link availability, and transfer AI / ML data to the UE in accordance with (e.g., based on) the mapping and priority comprised in the received AI / ML data configuration.

[0076] With reference to a second positioning example, an LMF may, for ML model training, want to provide a 10 GB dataset that comprises a fingerprinting dictionary that includes positioning-related measurements and ground truth locations to a UE. A fingerprinting dictionary may comprise a mapping between two parameters. For example, a fingerprinting dictionary may provide a mapping between physical channel measurement(s) and at least one UE location or whatever feature being estimated.

[0077] In the present example, 5 GB of the data comprises measurements repeated from the same locations. The LMF configures a gNB serving the UE to transfer data to the UE in two parts (of 5 GB each) using a priority order, so that UE first receives the measurements collected from different locations. This may enable better generalization performance, where generalization performance provides an indication of the performance how an ML model is maintained and / or varied when that machine learning model is deployed using new variables (e.g., a new environment and / or a new dataset). The generalization performance may be affected by differences in trained data and / or environment factors.

[0078] Further, for a management function (e.g., LMF, NWDAF) to make better (e.g., more accurate) decisions on data collection configuration, the management function may further obtain link quality metrics and / or link availability metrics from UEs and / or access nodes relating to past, current, or future (expected) instances. Stated differently, a management function may determine an AI / ML data configuration in dependence on received information indicating link availability and / or link quality for a link between the UE and the network. When an access node provides a management function with limited values and / or ranges of link quality, the management function may map only these values and / or ranges to data priorities.

[0079] Similarly, a management function may obtain information on content and / or characteristics of data available at the UE so as to prioritize different parts of it during a transfer. Stated differently, a management function may determine an AI / ML data configuration in dependence on received information indicating data content and / or data characteristics of data available at the UE.

[0080] Example signalling illustrating the presently described mechanisms is depicted in Figure 4.

[0081] Figure 4 illustrates signalling that may be performed between a UE 401, an access node 402, and a management function 403.

[0082] 4001 to 4003 are steps that are optionally performed. Stated differently, any (or none) of steps 4001 to 4003 may be performed prior to 4004.

[0083] During 4001, the UE 401 signals the management function 403. This signalling may comprise an indication of a link quality and / or a link availability between the UE 401 and the network node 402 (eg., a Uu link). The indication may relate to an uplink and / or downlink transmission direction. The indication of link quality may comprise, for example, at least one indication of uplink and / or downlink throughput statistics, current signal-to-interference and noise ratio (SINR) conditions, etc.

[0084] During 4002, the UE 401 signals the management function 403. This signalling may comprise an indication of information describing measurement data being signalled between the UE 401 and the access node 402 (e.g., a Uu link).

[0085] For example, the information describing measurement data may comprise at least one of: a type of data content comprised in the measurement data, a location associated with the measurement data, a time stamp associated with the measurement data, a positioning reference signal (or some other positioning signal) configuration used for obtaining the measurement data, or data statistics. In short, any information describing measurement data may be provided for use by a management function in prioritizing which data is to be provided for an AI / ML model.

[0086] Stated differently, in positioning mechanisms, an LMF acts as a hub for managing measurements and / or data collection for providing a location service and / or providing location information relating to an object. The LMF may first ask a data source (e.g., an access node and / or a UE) which data that data source has available and compares the indicated data to data already available to the LMF and / or data wanted by the LMF for providing the location service and / or providing the location information. This may be used by the LMF for determining what data to prioritize receiving from the data source.

[0087] During 4003, the access node signals the management function 403. This signalling may comprise an indication of a link quality and / or a link availability between the UE 401 and the access node 402 (e.g., a Uu link). The indication may relate to an uplink and / or downlink transmission direction. The indication of link quality may comprise, for example, at least one indication of uplink and / or downlink throughput statistics, current signal-to-interference and noise ratio (SINR) conditions, etc.

[0088] During 4004, the management function determines a configuration of data collection priority based on link conditions between the UE 401 and the access node 402 and the positioning Quality of Service (QoS) requirements to be met. The QoS requirements arise from positioning sessions (e.g., time durations during which a position of the UE 401 is to be determined) having associated requirements on reliability, integrity, and / or accuracy of positioning determination being made. The configuration is arranged to consider these QoS requirements to identify what data to be used for performing AI-ML mechanisms (also referred to herein as “AI / ML-related data”) has a relatively higher impact when obtaining a value for a position of the UE than others of those AI / ML-related data, and to designate those relatively higher impact AI / ML-related data as being of higher priority than the other AI / ML-related data.

[0089] For determining the data collection priority configuration, the managing entity may receive at least one indication of a link quality between the access node and the UE 401 from the UE 401 and / or the access node 402 during 4001 and / or 4003, and use this at least one indication to determine whether a prioritization of the transmittal of AI / ML-related data is to be performed.

[0090] The AI / ML-related data may comprise at least one or more of the following: * positioning-related measurements (e.g., time-, angle-, phase-, and / or powerbased measurements performed by the UE 401 (e.g., using a downlink PRS), and / or by the access node (e.g., using an uplink SRS, CIR, and / or PDP) that are input to the model being trained and / or run, * ground truth information associated with the measurements (such as UE location information, LOS / NLOS channel type indication, etc.) which are used as “labels” for training supervised ML models, or • parameters of the trained model, e.g., different layers of a neural network.

[0091] The data collection priority configuration may comprise an indication of which parts and / or characteristics of the AI-ML-related data is to be transferred at certain link conditions.

[0092] In all of the present examples, the data parts and / or characteristics may relate to one or more of the following properties of a given dataset: • Location, timestamp, positioning reference signal (PRS) configuration (e.g., transmission-reception points (TRPs) / anchors used, PRS resources, bandwidth, comb-size, etc.) associated with a (positioning-related) measurement / data » Statistics of the collected data: minimum (min), maximum (max), range, mean, variance, etc. In the embodiments, link quality / availability may relate to one or more of the following properties of a link between a UE and an access node (e.g., aUu link) in uplink and / or downlink: * (expected) min / max / range of throughput, latency, handover likelihood, handover information (source cell, candidate target cells, time, etc.), Uu radio measurements, e.g., reference signal received power (RSRP), reference signal received quality (RSRQ), etc.

[0093] Some of these options are illustrated by the following four examples.

[0094] In a first example, the AI-ML-related data may be partitioned into three sections; data section 1 (which is associated with a highest priority level); data section 2 (which is associated with a medium priority level); and data section 3 (which is associated with a lowest priority level.

[0095] In this second example, when link quality (e.g., throughput) is determined to be less than a minimum threshold (Z), then only data having a mean of X1 and variance Y1 is transmitted. When link quality is determined to be above the threshold Z, then data with a mean of X2 and variance of Y2 is transferred.

[0096] In this third example, when link quality (e.g., throughput) is determined to be less than a minimum threshold (Y) then only data collected from W locations (e.g., at least one (one or more) locations) is transferred. Wien link quality is determined to be above the threshold Y, then data collected from any location may be transferred.

[0097] In general, neural networks comprise multiple layers of neural networks, each layer processing data in a different manner. For some use cases, the last layer may be more useful for making estimations. In this fourth example, when link quality is determined to be less than the boundary threshold (Y), then only data corresponding to a last layer of a neural network is transferred. All data is transmitted when link quality is determined to be more than the threshold Y.

[0098] 4005 to 4007 relate to illustrating uplink data transfer, while 4008 to 4010 relate to illustrating downlink data transfer. It is therefore understood that 4004 may proceed to any of 4005 and / or 4008.

[0099] During 4005, the management function 403 signals the UE 401. This signalling may comprise a configuration that indicates how the UE 401 is to perform the data transfer to the access node 402 and / or the management function 403 with respect to the different priorities and link qualities. The UE 401 may apply the received configuration. It is understood that the UE 401 may have a known location (e.g., be a positioning reference unit), and / or have an unknown location (e.g., be an entity whose location is to be determined).

[0100] During 4006, the UE monitors link conditions between the UE 401 and the access node 402 by performing measurements thereon. The results of the measurements form a data collection. The UE may divide the data collection into a plurality of different sets according to the configuration received in 4005. Each set may correspond to at least one property of the measurement results comprised in that set and / or to at least one characteristic for how the measurement results comprised in that set were determined.

[0101] During 4007, the UE 401 signals the management function 403. This signalling may comprise at least one set of the plurality of sets of 4006, in accordance with the configuration received during 4005. For example, the UE 401 may determine that the configuration indicates that only a fraction (less than all) of the sets are to be transmitted based on a current and / or predicted link quality between the UE 401 and the access node 402, and transmit only that fraction of the sets. The remaining sets of the data collection may be transmitted at another time (e.g., when the UE 401 determines that the link quality has improved to a threshold amount), or deleted.

[0102] As mentioned above, 4008 to 4010 relate to downlink data transfers.

[0103] During 4008, the management function 403 signals the access node 402. This signalling may comprise a configuration that indicates how data collection to be reported by the access node 402 to the UE 401. The access node 402 may apply the received configuration. It is understood that the access node 402 may have a known location (e.g., be a positioning reference unit), and / or have an unknown location (e.g., be an entity whose location is to be determined).

[0104] During 4009, the access node 402 monitors link conditions between the UE 401 and the access node 402 by performing measurements thereon. The results of the measurements form a data collection. The access node 402 may divide the data collection into a plurality of different sets according to the configuration received in 4008. Each set may correspond to at least one property of the measurement results comprised in that set and / or to at least one characteristic for how the measurement results comprised in that set were determined.

[0105] During 4010, the access node 402 signals the UE 401. This signalling may comprise at least one set of the plurality of sets of 4009, in accordance with the configuration received during 4008. For example, the access node 402 may determine that the configuration indicates that only a fraction (less than all) of the sets are to be transmitted based on a current and / or predicted link quality between the UE 401 and the access node 402, and transmit only that fraction of the sets. The remaining sets of the data collection may be transmitted at another time (e.g., when the access node 402 determines that the link quality has improved to a threshold amount), or deleted.

[0106] In the example of FIG. 4, the communication between the management function and the UE may take place using any appropriate protocol, depending on the management function that is performing these actions. For example, when the management function is an LMF, the communications between the LMF and the UE may use, for example, a Long Time Evolution (LTE) Positioning Protocol (LPP) protocol, or New Radio positioning protocol a (NRPPa) and radio resource control (RRC) protocols. Similarly, communications between an access node and the management function may use any appropriate protocol. For example, when the access node is a gNB and the management function is an LMF, NRPPa protocol may be used.

[0107] Although the above examples are framed in the context of a measurement data being obtained between a UE and an access node, it is understood that the presently described techniques may be applied in respect of communications between other types of communication devices. This is illustrated with respect to FIG. 5, which illustrates applying the presently described techniques to sidelink communications between two UE.

[0108] FIG. 5 illustrates signalling that may be performed between a first UE 501, a second UE 502, and a management function 503 (e.g., an LMF and / or a server function, and / or a third UE).

[0109] 5001 to 5003 are steps that are optionally performed. Stated differently, any (or none) of steps 5001 to 5003 may be performed prior to 5004.

[0110] During 5001, the first UE 501 signals the management function 503. This signalling may comprise an indication of a link quality and / or a link availability between the first UE 501 and the second UE 502 (e.g., a sidelink). The indication may relate to a transmission direction. The indication of link quality may comprise, for example, at least one indication of throughput statistics, current signal-to-interference and noise ratio (SINR) conditions, etc.

[0111] During 5002, the first UE 501 signals the management function 503. This signalling may comprise an indication of a type of measurement data being signalled between the first UE 501 and the second UE 502.

[0112] During 5003, the second UE 502 signals the management function 503. This signalling may comprise an indication of a link quality and / or a link availability between the first UE 01 and the second UE 502. The indication may relate to a transmission direction. The indication of link quality may comprise, for example, at least one indication of throughput statistics, current signal-to-interference and noise ratio (SINR) conditions, etc. During 5004, the management function determines a configuration of data collection priority based on link conditions between the first UE 501 and the second UE 502 and the positioning Quality of Service (QoS) requirements to be met. The QoS requirements and the data collection priority configuration may be as described above in relation to FIG. 4. For example, For determining the data collection priority configuration, the management function may receive an indication of link quality between the first and second UEs (e.g., sidelink (SL) throughput statistics, and / or current SINR conditions, and / or SL congestion information (e.g., in terms of SL channel occupancy rate (CR), SL channel busy ratio (GBR), etc.), etc.).

[0113] During 5005, the management function 503 signals the first UE 501. This signalling may comprise a configuration that indicates how data collection to be reported by the UE 501 to the second UE 502 and / or the management function 503. The first UE 501 may apply the received configuration. It is understood that the first UE 501 and / or the second UE 502 may have a known location (e.g., be a positioning reference unit), and / or have an unknown location (e.g., be an entity whose location is to be determined).

[0114] During 5006, the first UE monitors link conditions between the first UE 501 and the second UE 502 by performing measurements thereon. The results of the measurements form a data collection. The UE may divide the data collection into a plurality of different sets according to the configuration received in 5005. Each set may correspond to at least one property of the measurement results comprised in that set and / or to at least one characteristic for how the measurement results comprised in that set were determined.

[0115] During 5007, the first UE 501 signals the second UE 502 and / or the management function 503. This signalling may comprise at least one set of the plurality of sets of 5006, in accordance with the configuration received during 5005. For example, the first UE 501 may determine that the configuration indicates that only a fraction (less than all) of the sets are to be transmitted based on a current and / or predicted link quality between the first UE 501 and the second UE 502, and transmit only that fraction of the sets. The remaining sets of the data collection may be transmitted at another time (e.g., when the first UE 501 determines that the link quality has improved to a threshold amount), or deleted.

[0116] The communication protocol used for communications between the first and second UE may use any appropriate communication protocol. For example, sidelink positioning protocol and / or sidelink Radio Resource Control (RRC) protocols may be used.

[0117] In an extension to the above examples, a management function for determining positioning and / or AI / ML data collection configurations (e.g., LMF, NWDAF or Server UE) may determine a configuration of AI / ML data collection priority based on link conditions of multiple links when the data being reported is transferred over these multiple links (e.g., for multiple hops). These multiple links may comprise any combination of uplink, downlink, and / or sidelink transmissions. In such a case, the management function may receive link quality information in respect of each of these multiple links. The management function subsequently provides the determined configuration to the involved apparatus, who transfer AI / ML data according to configuration they receive (as per the above examples of FIGs. 4 and 5).

[0118] FIGs. 4 and 5 illustrate features of the above examples. It is therefore understood that the above examples may find corresponding features in the below, and that the below apparatus may perform operations mentioned above in some examples.

[0119] FIG. 6 illustrates operations that may be performed by a first apparatus. The first apparatus may be comprised at or in an analytics function. For example, the apparatus may be comprised at or in a network data analytics function (NWDAF), an LMF, a UE (e.g., for sidelink operations), an access network node (e.g., a gNB), etc.

[0120] During 601, the first apparatus determines that data for use for a machine learning purpose is to be provided over a wireless link between a second apparatus and a third apparatus. The data may be for use in an ML model (e.g., for training and / or executing the ML model). The data may be used for ML functionality. The data may be used for an ML-enabled feature. The data may comprise measurement data. The data may comprise statistical data. The data may comprise labelled data and / or unlabelled data. The second and / or third apparatus may comprise at least one of an access node or a UE. For example, the second and the third apparatus may each comprise a UE, the second apparatus may comprise an access node and the third apparatus may comprise a UE (or vice versa), and / or the second and third apparatus may each comprise an access node (where those access nodes communicate using a link of variable quality, as mentioned below in relation to FIG. 7).

[0121] During 602, the first apparatus determines a transmission configuration for the second apparatus, wherein the transmission configuration associates a link quality of the wireless link with a prioritisation for transmitting the data. The transmission configuration may associate a link quality for data to be transmitted by the second apparatus over the wireless link with the prioritization for transmitting the data. The prioritization may be based on at least one of a plurality of different factors. For example, the prioritization may be based on a characteristic of the data (e.g., content, importance, etc.). The transmission configuration may comprise at least one of: uplink transmission, downlink transmission, or sidelink transmission (e.g., transmission between peer apparatus).

[0122] During 603, the first apparatus provides the transmission configuration to the second apparatus. For example, the first apparatus may transmit the transmission configuration to the second apparatus.

[0123] The first apparatus may perform the same operations in respect of the third apparatus as for the second apparatus.

[0124] The determining the transmission configuration may comprise: determining a first threshold of the link quality; and associating the first threshold with a first portion of data with a first priority to be reported by the second apparatus, wherein the first portion of data is less than all data to be transmitted during a first transmission opportunity. Stated differently, the first threshold and an identification of the first portion of data associated with a first priority for transmission may be comprised in the transmission configuration.

[0125] The determining the transmission configuration may comprise: determining a second threshold of the link quality, wherein the second threshold is different from (e.g., higher than) the first threshold; and associating the second threshold with a second portion of data with a second priority to be reported by the second apparatus, wherein the second portion of data is less than all data to be transmitted during the first transmission opportunity. Stated differently, the second threshold and an identification of the second portion of data associated with a second priority may be comprised in the transmission configuration. Similarly, a third threshold and an identification of a third portion of data associated with a third priority may be comprised in the transmission configuration when there is provided a third threshold of the link quality that is different from (e.g., higher than) the first and second thresholds.

[0126] The determining the transmission configuration may comprise: determining a second threshold of the link quality, wherein the second threshold is different from (e.g., higher than) the first threshold; and associating the second threshold with a second portion of data with a second priority to be reported by the second apparatus, wherein the second portion of data comprises all data to be transmitted during the first transmission opportunity. Stated differently, the second threshold and an identification of the second portion of data associated with a second priority may be comprised in the transmission configuration.

[0127] The first apparatus may determine the prioritisation for transmitting the data, wherein determining the prioritisation is based on at least one of: an identification of a portion of said data that is to be used for the machine learning purpose but which is unavailable at a receiver of the data; an identification of a portion of the data that comprises a lower redundancy relative to other data to be transmitted; an identification of a portion of the data that can be used for multiple purposes for the machine learning purpose; or an identification of a portion of the data fulfilling a minimum performance metric.

[0128] The performance metric may comprise at least one of: a quality of service of data transmission; an accuracy of data transmission; or an integrity of data transmission.

[0129] The determining the transmission configuration may comprise: mapping different portions of the data to be transmitted by the second apparatus to respective transmission link conditions, the respective link conditions being indicated by at least one respective threshold or respective range. The mapping may be comprised in the transmission configuration.

[0130] FIG. 7 illustrates operations that may be performed by a second apparatus. The second apparatus may comprise the second apparatus discussed above in relation to FIG. 6. The second apparatus may comprise an access node (e.g., a gNB). The second apparatus may comprise a UE. The second apparatus may comprise an apparatus configured to transmit over a transmission link having a dynamically varying quality (e.g., an over-the-air link). The first apparatus referenced below may comprise the first apparatus mentioned in relation to FIG. 6.

[0131] During 701, the second apparatus receives, from a first apparatus, a transmission configuration, wherein the transmission configuration associates a link quality of a wireless link between the second apparatus and a third apparatus with a prioritisation for transmitting the data. The data, transmission configuration, and / or prioritization may be as discussed above in relation to FIG. 6.

[0132] During 702, the second apparatus determines a link quality of the wireless link. Stated differently, the second apparatus may determine a link quality for data currently being transmitted over the wireless link from the second apparatus to the third apparatus.

[0133] During 703, the second apparatus selects a portion of the data to be transmitted over the wireless link using the transmission configuration. For example, when the link quality is determined to be relatively bad, the selected portion is smaller than the portion selected when the link quality is determined to be relatively good. The selection of the data may be performed using the prioritization comprised in the transmission configuration.

[0134] During 704, the second apparatus transmits the selected portion to the third apparatus over the wireless link. This transmission may be in accordance with the received transmission configuration. Stated differently, this transmission may be in accordance with the prioritization comprised in the transmission configuration.

[0135] The received transmission configuration may be as described above in relation to FIG. 6.

[0136] The transmission configuration may comprise: a first threshold of the link quality associated with a first portion of data associated with a first priority to be reported by the second apparatus. The first portion of data may be less than all data to be transmitted during a first transmission opportunity.

[0137] The transmission configuration may comprise: a second threshold of the link quality associated with a second portion of data associated with a second priority to be reported by the second apparatus. The second portion of data may be less than all data to be transmitted during the first transmission opportunity.

[0138] The transmission configuration may comprise: a second threshold of the link quality associated with a second portion of data associated with a second priority to be reported by the second apparatus. The second portion of data may comprise all data to be transmitted during the first transmission opportunity.

[0139] The selecting the portion of data may be performed using the prioritization comprised in the transmission configuration. For example, the selecting the portion of data may be performed in dependence on at least one association between a link quality (provided, for example, as a threshold and / or a range of values) and an identification of data to be transmitted when a current link quality falls within the values associated with that link quality.

[0140] Stated differently, the selecting the portion of data may be based on at least one of: an identification of a portion of said data that is to be used for a machine learning purpose but which is unavailable at a receiver of the data; an identification of a portion of the data that comprising a lower redundancy relative to other data to be transmitted; an identification of a portion of the data that can be used for multiple purposes for the machine learning purpose; or an identification of a portion of data fulfilling a minimum performance metric.

[0141] The performance metric may comprise at least one of: a quality of service; an accuracy; or an integrity.

[0142] The transmission configuration may comprise a mapping that maps different portions of the data to be transmitted by the second apparatus to respective transmission link conditions, the respective link conditions being indicated by at least one respective threshold or respective range.

[0143] The foregoing description has provided by way of non-limiting examples a full and informative description of some examples. However, various modifications and adaptations may become apparent to those skilled in the relevant arts in view of the foregoing description, when read in conjunction with the accompanying drawings and the claims. However, all such and similar modifications of the teachings will still fall within the scope of the claims.

[0144] In the above, different examples are described using, as an example of an access architecture to which the described techniques may be applied, a radio access architecture based on long term evolution advanced (LTE Advanced, LTE-A) or new radio (NR, 5G), without restricting the examples to such an architecture, however. The examples may also be applied to other kinds of communications networks having suitable means by adjusting parameters and procedures appropriately. Some examples of other options for suitable systems are the universal mobile telecommunications system (UMTS) radio access network (UTRAN), wireless local area network (WLAN or Wi-Fi), worldwide interoperability for microwave access (WiMAX), Bluetooth®, personal communications services (PCS), ZigBee®, wideband code division multiple access (WCDMA), systems using ultra-wideband (UWB) technology, sensor networks, mobile ad-hoc networks (MANETs) and Internet Protocol multimedia subsystems (SMS) or any combination thereof.

[0145] As provided herein, various aspects are described in the detailed description of examples and in the claims. In general, some examples may be implemented in hardware or special purpose circuits, software code, logic or any combination thereof. For example, some aspects may be implemented in hardware, while other aspects may be implemented in firmware or software code which may be executed by a controller, microprocessor or other computing device, although examples are not limited thereto. While various examples may be illustrated and described as block diagrams, flow charts, or using some other pictorial representation, it is well understood that these blocks, apparatus, systems, techniques or methods described herein may be implemented in, as non-limiting examples, hardware, software code, firmware code, special purpose circuits or logic, general purpose hardware or controller or other computing devices, or some combination thereof.

[0146] The examples may be implemented by computer software code stored in a memory and executable by at least one data processor of the involved entities or by hardware, or by a combination of software code and hardware.

[0147] The memory referred to herein may be of any type suitable to the local technical environment and may be implemented using any suitable data storage technology, such as semiconductor based memory devices, magnetic memory devices and systems, optical memory devices and systems, fixed memory and removable memory.

[0148] The (data) processors referred to herein may be of any type suitable to the local technical environment, and may comprise one or more of general purpose computers, special purpose computers, microprocessors, digital signal processors (DSPs), application specific integrated circuits (ASIC), FPGA, gate level circuits and processors based on multi core processor architecture, as non-limiting examples.

[0149] Further in this regard it should be noted that any procedures, e.g.. as in FIG. 6, and / or FIG. 7, and / or otherwise described previously, may represent operations of a computer program being deployed by at least one processor comprised in an apparatus (where a computer program comprises instructions for causing an apparatus to perform at least one action, the instructions being represented as software code stored on at least one memory), or interconnected logic circuits, blocks and functions, or a combination of operations of a computer program being deployed by at least one processor comprised in an apparatus and logic circuits, blocks and functions. The software code may be stored on memory, such as physical media as memory chips, or memory blocks implemented within the processor, magnetic media (such as hard disk or floppy disks), and optical media (such as for example DVD and the data variants thereof, CD, and so forth).

[0150] The memory may be of any type suitable to the local technical environment and may be implemented using any suitable data storage technology, such as semiconductor-based memory devices, magnetic memory devices and systems, optical memory devices and systems, fixed memory and removable memory. The data processors may be of any type suitable to the local technical environment, and may include one or more of general purpose computers, special purpose computers, microprocessors, digital signal processors (DSPs), application specific integrated circuits (ASIC), gate level circuits and processors based on multicore processor architecture, as nonlimiting examples.

[0151] Additionally or alternatively, some examples may be implemented using circuitry. The circuitry may be configured to perform one or more of the functions and / or method steps previously described. That circuitry may be provided in the base station and / or in the communications device and / or in a core network entity.

[0152] As used in this application, the term “circuitry” or “means” may refer to one or more or all of the following: (a) hardware-only circuit implementations (such as implementations in only analogue and / or digital circuitry); (b) combinations of hardware circuits and software cade, such as: (i) a combination of analogue and / or digital hardware circuit(s) with software / firmware code and (ii) any portions of hardware processor(s) with software code (including digital signal processor(s)), software code, and memory(ies) that work together to cause an apparatus, such as the communications device or base station to perform the various functions previously described; and (c) hardware circuit(s) and or processor(s), such as a microprocessor(s) or a portion of a microprocessor(s), that requires software code (e.g., firmware) for operation, but the software code may not be present when it is not needed for operation.

[0153] 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 code. The term circuitry also covers, for example integrated device.

[0154] Implementations of the disclosure may be practiced in various components such as integrated circuit modules. The design of integrated circuits is by and large a highly automated process. Complex and powerful software tools are available for converting a logic level design into a semiconductor circuit design ready to be etched and formed on a semiconductor substrate.

[0155] 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.

[0156] 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).

[0157] The scope of protection sought for various examples of the disclosure is set out by the independent claims. The examples and features, if any, described in this specification that do not fall under the scope of the independent claims are to be interpreted as examples useful for understanding the disclosure.

[0158] The foregoing description has provided by way of non-limiting examples a full and informative description of example implementations of this disclosure. However, various modifications and adaptations may become apparent to those skilled in the relevant arts in view of the foregoing description, when read in conjunction with the accompanying drawings and the appended claims. However, all such and similar modifications of the teachings of this disclosure will still fall within the scope of this invention as defined in the appended claims. Indeed, there is a further implementation comprising a combination of one or more implementations with any of the other implementations previously discussed.

Claims

1) A first apparatus comprising means for performing:determining that data for use for a machine learning purpose is to be provided over a wireless link between a second apparatus and a third apparatus;determining a transmission configuration for the second apparatus, wherein the transmission configuration associates a link quality of the wireless link with a prioritisation for transmitting the data,; andproviding the transmission configuration to the second apparatus.2) A first apparatus as claimed in claim 1, wherein determining the transmission configuration comprises:determining a first threshold of the link quality; andassociating the first threshold with a first portion of data with a first priority to be reported by the second apparatus, wherein the first portion of data is less than all data.3) A first apparatus as claimed in claim 2, wherein determining the transmission configuration comprises:determining a second threshold of the link quality, wherein the second threshold is different from the first threshold; andassociating the second threshold with a second portion of data with a second priority to be reported by the second apparatus, wherein the second portion of data is less than all data and is different from the first portion of data.4) A first apparatus as claimed in claim 2, wherein determining the transmission configuration comprises:determining a second threshold of the link quality, wherein the second threshold is different from the first threshold; andassociating the second threshold with a second portion of data with a second priority to be reported by the second apparatus, wherein the second portion of data comprises all data.5) A first apparatus as claimed in any preceding claim, further comprising means for determining the prioritisation for transmitting the data, wherein determining the prioritisation is based on at least one of: an identification of a portion of said data that is to be used for the machine learning purpose but which is unavailable at a receiver of the data; an identification of a portion of the data that comprises a lower redundancy relative to other data to be transmitted; an identification of a portion of the data that can be used for multiple purposes for the machine learning purpose; or an identification of a portion of the data fulfilling a minimum performance metric.6) A first apparatus as claimed in claim 5, wherein the performance metric comprises at least one of: a quality of service; an accuracy; or an integrity.7) A first apparatus as claimed in any preceding claim, wherein determining the transmission configuration comprises: mapping different portions of the data to be transmitted by the second apparatus to respective transmission link conditions, the respective link conditions being indicated by at least one respective threshold or respective range.8) A second apparatus comprising means for performing:receiving, from a first apparatus, a transmission configuration, wherein the transmission configuration associates a link quality of a wireless link between the second apparatus and a third apparatus with a prioritisation for transmitting the data;determining a link quality of the wireless link;selecting a portion of the data to be transmitted over the wireless link using the transmission configuration; andtransmitting the selected portion to the third apparatus over the wireless link.9) A second apparatus as claimed in claim 8, wherein the transmission configuration comprises:a first threshold of the link quality associated with a first portion of data associated with a first priority to be reported by the second apparatus, wherein the first portion of data is less than all data.10)A second apparatus as claimed in claim 8, wherein the transmission configuration comprises:a second threshold of the link quality associated with a second portion of data associated with a second priority to be reported by the second apparatus, wherein the second portion of data is less than all data.11)A second apparatus as claimed in claim 8, wherein the transmission configuration comprises:a second threshold of the link quality associated with a second portion of data associated with a second priority to be reported by the second apparatus, wherein the second portion of data comprises all data.12)A second apparatus as claimed in any of claims 8 to 11, wherein the selecting the portion of data is based on at least one of: an identification of a portion of said data that is to be used for a machine learning purpose but which is unavailable at a receiver of the data; an identification of a portion of the data that comprising a lower redundancy relative to other data to be transmitted; an identification of a portion of the data that can be used for multiple purposes for the machine learning purpose; or an identification of a portion of the data fulfilling a minimum performance metric.13)A second apparatus as claimed in claim 12, wherein the performance metric comprises at least one of: a quality of service; an accuracy; or an integrity.14)A second apparatus as claimed in any of claims 8 to 13, wherein the transmission configuration comprises a mapping that maps different portions of the data to be transmitted by the second apparatus to respective transmission link conditions, the respective link conditions being indicated by at least one respective threshold or respective range.15)An apparatus as claimed in any preceding claim, wherein the first apparatus is comprised in a location management function and / or a network data analytics function and / or an access node and / or a user equipment.16)An apparatus as claimed in any preceding claim, wherein the second apparatus and / or third apparatus is comprised in an access node and / or a user equipment.17)A method for a first apparatus, the method comprising:determining that data for use for a machine learning purpose is to be provided over a wireless link between a second apparatus and a third apparatus;determining a transmission configuration for the second apparatus, wherein the transmission configuration associates a link quality of the wireless link with a prioritisation for transmitting the data; andproviding the transmission configuration to the second apparatus.18)A method for a second apparatus, the method comprising:receiving, from a first apparatus, a transmission configuration, wherein the transmission configuration associates a link quality of a wireless link between the second apparatus and a third apparatus with a prioritisation for transmitting the data;determining a link quality of the wireless link;selecting a portion of the data to be transmitted over the wireless link using the transmission configuration; andtransmitting the selected portion to the third apparatus over the wireless link.19)A computer program comprising instructions which, when executed by a first apparatus, cause the first apparatus at least to perform:determining that data for use for a machine learning purpose is to be provided over a wireless link between a second apparatus and a third apparatus;determining a transmission configuration for the second apparatus, wherein the transmission configuration associates a link quality of the wireless link with a prioritisation for transmitting the data; andproviding the transmission configuration to the second apparatus.20)A computer program comprising instructions which, when executed by an second apparatus, cause the second apparatus at least to perform:receiving, from a first apparatus, a transmission configuration, wherein the transmission configuration associates a link quality of a wireless link between the second apparatus and a third apparatus with a prioritisation for transmitting the data;determining a link quality of the wireless link;selecting a portion of the data to be transmitted over the wireless link using the transmission configuration; andtransmitting the selected portion to the third apparatus over the wireless link.

Images