User equipment, network node and method

Abstract

The present disclosure relates to user equipment, network nodes, and methods. A user equipment, UE, comprising at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the UE to at least perform: receiving, from a network node, configuration information associated with a data collection request, the data collection request being transmitted by the UE and being associated with one or more UE-side machine learning models, wherein the configuration information indicates a permission condition for the data collection request; and transmitting, to the network node, a request for data collection based on the indicated permission condition for the data collection request.

Description

Cross-references to related applications

[0001] This application claims priority and benefit to EP patent application No. 25166838.0, filed on March 27, 2025, the contents of which are incorporated herein by reference in their entirety. Technical Field

[0002] Examples disclosed herein relate to user equipment, network nodes, and methods. Some relate to user equipment, network nodes, and methods for data collection. Background Technology

[0003] A user equipment (UE) may send a request for data collection to one or more network functions. The request for data collection may request data associated with channel measurements and / or truth labels associated with channel measurements. Summary of the Invention

[0004] According to various, but not all, examples, a user equipment (UE) is provided, comprising: at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the UE to at least: receive configuration information associated with a data collection request sent by the UE and associated with one or more UE-side machine learning models from a network node, wherein the configuration information indicates permission conditions for the data collection request; and send a data collection request to the network node based on the indicated permission conditions for the data collection request.

[0005] In some, but not all, examples, the received configuration information indicates one or more of the following: one or more network areas for which a data collection permission request is made; one or more time periods for which a data collection permission request is made; one or more reference signal configurations for which a data collection permission request is made; one or more measurement and reporting configurations for which a data collection permission request is made; the maximum amount of data that can be requested; one or more permission purposes for which data can be requested; one or more device types for which a data collection permission request is made; or one or more vendors for which a data collection permission request is made.

[0006] In some, but not all, examples, the configuration information also indicates the information that needs to be included in the data collection request.

[0007] In some, but not all, examples, the information that needs to be included in the data collection request includes one or more of the following: information associated with the UE’s supplier; information associated with the purpose of data collection; or an indication of the amount of data requested.

[0008] In some, but not all, examples, the instructions, when executed by at least one processor, also cause the UE to: determine whether one or more machine learning models require data collection; and, based on the determination that one or more machine learning models require data collection, send a request for data collection to a network node based on configuration information.

[0009] In some, but not all, examples, when the instructions are executed by at least one processor, the UE also performs the following: determining whether one or more machine learning models require data collection; and, based on the determination that one or more machine learning models require data collection, sending a request to a network node for configuration information associated with the data collection request sent by the UE and associated with one or more UE-side machine learning models.

[0010] In some, but not all, examples, the instructions, when executed by at least one processor, also cause the UE to: receive data from a network node associated with a sent request for data collection; and one or more of the following: train one or more machine learning models in the machine learning models using the received data; monitor one or more machine learning models in the machine learning models using the received data; fine-tune one or more machine learning models in the machine learning models using the received data; and perform inference using one or more machine learning models in the machine learning models and the received data.

[0011] In some, but not all, examples, data collection requests include requests for the collection of data associated with one or more of the following: channel measurements; or truth labels.

[0012] In some, but not all, examples, the instructions, when executed by at least one processor, also cause the UE to perform: use one or more machine learning models from the machine learning models for one or more of the following: positioning; beam management; mobility; resource allocation; data transmission; or data reception.

[0013] According to various, but not necessarily all, examples, a method is provided, comprising: receiving configuration information associated with a data collection request from a network node, the data collection request being sent by a UE and associated with one or more UE-side machine learning models, wherein the configuration information indicates permission conditions for the data collection request; and sending a data collection request to the network node based on the indicated permission conditions for the data collection request.

[0014] According to various, but not all, examples, a network node is provided, comprising: at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the network node to perform at least: determining configuration information associated with a data collection request sent by a user equipment (UE) and associated with one or more UE-side machine learning models, wherein the configuration information indicates permission conditions for the data collection request; sending the configuration information to the UE; and receiving the data collection request from the UE.

[0015] In some, but not all, examples, the configuration information indicates one or more of the following: one or more network areas for which a data collection request is permitted; one or more time periods for which a data collection request is permitted; one or more reference signal configurations for which a data collection request is permitted; one or more measurement and reporting configurations for which a data collection request is permitted; the maximum amount of data that can be requested; one or more permitted purposes for which data can be requested; one or more device types for which a data collection request is permitted; or one or more vendors for which a data collection request is permitted.

[0016] In some, but not all, examples, determining configuration information includes one or more of the following: determining the data load at at least one location in at least one time period, and determining one or more network areas and / or one or more time periods for which permission to request data collection is granted based on the determined data load; determining the size of the dataset maintained by the network nodes, and determining the maximum amount of data that can be requested based on the determined size of the dataset; Determine the number of past data collection requests received from at least one device and / or at least one vendor, and determine one or more devices and / or one or more device vendors that authorize the data collection requests based on the number of past data collection requests received.

[0017] In some, but not all, examples, the instructions, when executed by at least one processor, also cause the network node to: determine the data to be sent to the UE based on the received data collection request; and send the determined data to the UE.

[0018] Based on various, but not necessarily all, examples, a method is provided that includes: determining configuration information associated with a data collection request, the data collection request being sent by a user equipment (UE) and associated with one or more UE-side machine learning models, wherein the configuration information indicates permission conditions for the data collection request; sending the configuration information to the UE; and receiving the data collection request from the UE.

[0019] According to various, but not necessarily all, embodiments, an apparatus is provided, comprising: at least one processor; and at least one memory; the at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus to perform at least a portion of one or more methods described herein.

[0020] According to various, but not necessarily all, embodiments, an apparatus is provided including components for performing at least a portion of one or more methods described herein. The description of functions and / or actions should also be considered as disclosing any components suitable for performing those functions and / or actions. The functions and / or actions described herein can be performed using any suitable method and in any suitable manner.

[0021] Examples as claimed in the appended claims are provided according to various, but not necessarily all, embodiments.

[0022] While the examples and optional features of this disclosure have been described separately, it should be understood that all possible combinations and permutations thereof are included within this disclosure. It should be understood that various examples of this disclosure may include any or all of the features described with respect to other examples of this disclosure, and vice versa. Furthermore, it should be understood that any one or more features in any combination may be implemented / included / performed by means of, as needed and appropriately, by means of, apparatus, method, and / or instructions. The description of a function should also be considered as disclosing any components suitable for performing that function. Attached Figure Description

[0023] Some examples will now be described with reference to the accompanying drawings, in which: Figure 1 Examples of the topics described in this article are shown; Figure 2 Another example of the topic described in this article is shown; Figure 3 Another example of the topic described in this article is shown; Figure 4A Another example of the topic described in this article is shown; Figure 4B Another example of the topic described in this article is shown; Figure 5 Another example of the topic described in this article is shown; Figure 6 Another example of the subject described in this article is shown; and Figure 7 Another example of the topic described in this article is shown.

[0024] The accompanying drawings are not necessarily drawn to scale. For clarity and brevity, some features and views in the drawings may be shown schematically or enlarged to scale. For example, the dimensions of some elements in the drawings may be enlarged relative to other elements to aid illustration. Similar reference numerals are used in the drawings to indicate similar features. For clarity, not all reference numerals may be shown in all drawings. Detailed Implementation

[0025] The accompanying drawing illustrates a user equipment (UE) including: at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the UE to at least: receive configuration information associated with a data collection request from a network node, the data collection request being sent by the UE and associated with one or more UE-side machine learning models, wherein the configuration information indicates permission conditions for the data collection request; and send a data collection request to the network node based on the indicated permission conditions for the data collection request.

[0026] This provides the following technical effects: controlling UE requests for data collection, suppressing data collection requests that consume excessive radio resources, redundant data requests, and overly demanding data requests.

[0027] The UE may request data collection from network functions and / or from network nodes (e.g., location management function (LMF)) for use by a UE-side artificial intelligence (AI) / machine learning (ML) model (referred to as the "ML model" for clarity). Network functions may be performed by network nodes.

[0028] The UE-side ML model can be utilized by the UE or by another UE-side node or function (e.g., the UE-side server).

[0029] In some, but not all, examples, one or more ML models are used for one or more of the following: positioning; beam management; mobility; resource allocation; data transmission; or data reception.

[0030] Without controlling UE requests for data collection, a UE could send a data collection request to the network node at any time. This could result in a high frequency of data collection requests, potentially consuming significant radio resources. Furthermore, depending on the processing load at the network node, it might be unable to process and respond to all received data collection requests.

[0031] Furthermore, if UE requests for data collection are not controlled, redundant data collection requests may be sent to network nodes. This could happen, for example, if a vendor instructs all its UEs to send the same or similar data collection requests to network nodes simultaneously and at a high frequency.

[0032] Furthermore, if UE requests for data collection are not controlled, the UE may send data collection requests that the network node cannot fulfill. For example, this could happen if the UE requests a larger amount of data than the network node is capable of sending.

[0033] The mechanism for initiating data collection from the ML model on the UE side controlled by the network is described below.

[0034] Figure 1 An example of network 100 is shown, which includes multiple network entities, including terminal device 110, node device 120, and one or more network devices 130. Terminal device 110 and node device 120 communicate with each other 124. One or more network devices 130 communicate with node device 120 128.

[0035] In some examples, one or more network devices 130 communicate with terminal device 110. In some examples, one or more network devices 130 may communicate with each other. In some examples, one or more node devices 120 may communicate with each other.

[0036] Network 100 may be a cellular network comprising multiple cells 122, each cell being served by node device 120. In this example, the interface between terminal device 110 and node device 120 defining cell 122 is wireless interface 124.

[0037] Node device 120 includes one or more cellular radio transceivers. Terminal device 110 includes one or more cellular radio transceivers.

[0038] In the example shown, cellular network 100 is a 3GPP network, where terminal device 110 is a user equipment (UE), and node device 120 may be an access node such as a base station.

[0039] User equipment includes mobile devices. Where references to user equipment are made, such references include and cover references to mobile devices, where possible.

[0040] In some examples, during operation, user equipment 110 includes a mobile device that includes a smart card, such as a subscriber identity module (SIM), for authentication / encryption, etc. In some examples, during operation, user equipment 110 includes a mobile device that includes circuitry for authentication / encryption, such as a software SIM, embedded as part of user equipment 110.

[0041] Node device 120 can be any suitable access node, such as a base station or transmit / receive point. Node device 120 can be a network element responsible for radio transmission and reception to or from UE 110 in one or more cells 122. Node device 120 can be a network element in a radio access network (RAN), an open radio access network (O-RAN), or any other suitable type of network.

[0042] Network device 130 may be part of the core network. Network device 130 may be configured to manage functions related to connectivity with UE 110. For example, network device 130 may be configured to manage functions such as connectivity, mobility, authentication, authorization, and / or other suitable functions. In some examples, network device 130 may include Access and Mobility Management Functions (AMF) and / or User Plane Functions (UPF) or any other suitable entity.

[0043] exist Figure 1 In the example, network device 130 is shown as a single entity. In some examples, network device 130 may be distributed across multiple entities. For example, network device 130 may be cloud-based or distributed in any other suitable manner. Network device 130 may be a core network node.

[0044] For example, network 100 can be a 4G or 5G network. For instance, it can be a New Radio (NR) network using a gNB or eNB as access node 120. New Radio is the 3GPP name for 5G technology. In this case, node device 120 can include a gNodeB (gNB) 120 configured to provide user plane and control plane protocol termination and / or perform any other suitable functions to UE 110. gNBs 120 interconnect with each other via X2 / Xn interface 126. gNBs are also connected to network device 130 via N2 interface 128. gNBs can connect to an AMF or any other suitable network device 130. Other types of networks and interfaces can be used in other examples. Other types of networks can include next-generation mobile and communication networks, such as 6G networks.

[0045] Figure 2 An example of method 200 is shown.

[0046] In the example, it can be considered that Figure 2 Several methods are shown. For example, Figure 2 It shows one or more actions at multiple participants / entities, and in the example, Figure 2 This can be seen as demonstrating multiple methods performed by the various participants / entities.

[0047] about Figure 2One or more features discussed may appear in one or more other figures.

[0048] exist Figure 2 In the example, multiple devices send and / or receive one or more signals and / or one or more messages via and / or using a network. In the example, any suitable form of communication can be used from any suitable network. For example, it can be used... Figure 1 At least a portion of the network 100.

[0049] Therefore, in the example, Figure 2 Multiple devices in the middle form as about Figure 1 At least a portion of the network 100 described.

[0050] In the example shown, terminal node 110 and access node 120 send and / or receive one or more signals and / or one or more messages. The access node may include a gNodeB (gNB), and terminal node 110 may include a UE.

[0051] In the example, Figure 2 Communication and / or transmission between the shown components can be carried out via any number of intermediate components, including without intermediate components.

[0052] Despite Figure 2 The example shows one terminal node 110, but in this example, any suitable number of terminal nodes 110, such as UEs, may be included. Similarly, in this example, any suitable number of access nodes 120 may be included.

[0053] As described herein, descriptions of functions and / or actions should also be considered as disclosing enabling and / or causing and / or controlling such functions and / or actions. For example, descriptions of sending information should also be considered as disclosing enabling and / or causing and / or controlling the sending / transmission of information.

[0054] For example, a description of a device for transmitting information (such as a UE) should also be considered as disclosing that at least one controller of the device enables and / or causes and / or controls the device to transmit information. In the example shown, the position of the box indicates the entity performing the function and / or action.

[0055] because Figure 2 It shows one or more actions / features that were sent, so Figure 2 The corresponding receive / enable and / or receive (multiple) actions / features are shown.

[0056] for Figure 2Further discussion will consider terminal node 110 as a UE and access node 120 as a network node providing LMF. In other examples, the access node may be a network node providing access management function (AMF), a network node providing operation and management function (OMF), or a gNB.

[0057] Signaling between a terminal node and an access node can occur via one or more signaling protocols. Such signaling protocols may include the LTE Positioning Protocol (LPP); Radio Resource Control (RRC); or Media Access Control (MAC).

[0058] From the UE's perspective, the method includes, at box 202, determining whether one or more UE-side machine learning models require data collection.

[0059] In some, but not all, examples, determining whether one or more ML models require data collection includes determining that at least one ML model requires one or more of the following: training, monitoring, or fine-tuning. In some, but not all, examples, determining whether one or more ML models require data collection includes determining that the dataset currently available for use with one or more ML models does not contain sufficient data to perform at least one of the following: training, monitoring, or fine-tuning the ML model.

[0060] In some, but not all, examples, determining whether data collection is required for one or more ML models includes determining the requirements for performing inference using one or more ML models. In some, but not all, examples, determining whether data collection is required for one or more ML models includes determining that the dataset currently available for use with one or more ML models does not contain sufficient data to perform inference using one or more ML models.

[0061] At box 204, the method includes sending a request to a network node for configuration information associated with a data collection request sent by the UE and associated with one or more UE-side machine learning models (referred to herein as "data collection configuration information").

[0062] Based on the determination that one or more machine learning models require data collection to perform transmission to network nodes.

[0063] In some, but not all, examples, sending a request for data collection configuration information before sending a request for data collection in general allows the UE to know the restrictions imposed by the network node on the data collection request. This avoids unnecessary resource usage, such as sending data collection requests that are likely to be rejected.

[0064] In some, but not all, examples, the data collection configuration information indicates one or more space-based restrictions on the UE's data collection request. In some, but not all, examples, the data collection configuration information indicates one or more network areas where the data collection request is permitted. In some, but not all, examples, the data collection configuration information indicates one or more network areas where the data collection request is not permitted.

[0065] In some, but not all, examples, indications for one or more network areas requesting permitted or unpermitted data include indications for one or more of the following: geographic area; tracking area; RAN notification area (RNA); serving cell / gNB. In some, but not all, examples, indications for one or more network areas requesting permitted or unpermitted data include one or more of the following: geographic area; tracking area; RNA; serving cell / gNB.

[0066] In some, but not all, examples, the data collection configuration information indicates one or more time-based restrictions on the UE's data collection request. In some, but not all, examples, the data collection configuration information indicates one or more time periods during which the data collection request is permitted. In some, but not all, examples, the data collection configuration information indicates one or more time periods during which the data collection request is not permitted.

[0067] In some, but not all, examples, indications for one or more time periods of permitted or denied data requests include indications for one or more of the following: time of day; date; or time window.

[0068] In some, but not all, examples, the data collection configuration information indicates one or more configuration-based restrictions on the UE's data collection request.

[0069] In some, but not all, examples, the data collection configuration information indicates one or more reference signal configurations that permit data requests. In some, but not all, examples, the data collection configuration information indicates one or more reference signal configurations that do not permit data requests.

[0070] In some, but not all, examples, the data collection configuration information indicates one or more measurement and reporting configurations that permit data requests. In some, but not all, examples, the data collection configuration information indicates one or more measurement and reporting configurations that do not permit data requests.

[0071] One or more configuration-based restrictions on a UE data collection request indicate that the UE may request data associated with certain permitted configurations and / or permitted network-side conditions. In some, but not all, examples, the permitted configurations and / or permitted network-side conditions are identified using associated identifiers (IDs). In some, but not all, examples, the permitted configurations indicate the configurations that the UE may send the data collection request for. Example configurations include area information; time information; and permitted data volume. In some, but not all, examples, one or more configuration-based restrictions on a UE data collection request indicate that the UE may request data from the network that meets the permitted configuration criteria.

[0072] In some, but not all, examples, one or more configuration-based restrictions on the UE data collection request indicate one or more of the following: the maximum number of configurations; the range of associated IDs for determining the configurations; and parameters for one or more configurations used to collect data.

[0073] In some, but not all, examples, the data collection configuration information indicates one or more data volume-based limits on the UE's data collection requests. In some, but not all, examples, the data collection configuration information indicates the maximum amount of data that can be requested.

[0074] In some, but not all, examples, the maximum amount of data that can be requested indicates the maximum number of data samples that can be requested. In some, but not all, examples, for one or more individual data types, the maximum amount of data that can be requested indicates the maximum number of data samples of that individual data type that can be requested.

[0075] In some, but not all, examples, the data collection configuration information indicates one or more purpose-based restrictions on the UE's data collection request. In some, but not all, examples, the data collection configuration information indicates one or more permitted purposes for which data can be requested. In some, but not all, examples, the data collection configuration information indicates one or more prohibited purposes for which data cannot be requested.

[0076] In some, but not all, examples, one or more permitted or unpermitted purposes include one or more of the following: training an ML model; monitoring an ML model; fine-tuning an ML model; or using an ML model to perform inference.

[0077] In some, but not all, examples, the data collection configuration information indicates one or more device-based restrictions on the UE's data collection request. In some, but not all, examples, the data collection configuration information indicates one or more device types and / or device capabilities that permit the data collection request. In some, but not all, examples, the data collection configuration information indicates one or more device types and / or device capabilities that do not permit the data collection request.

[0078] In some, but not all, examples, the data collection configuration information indicates one or more vendor-based restrictions on the UE's data collection request. In some, but not all, examples, the data collection configuration information indicates one or more vendors that permit the data collection request. In some, but not all, examples, the data collection configuration information indicates one or more vendors that do not permit the data collection request.

[0079] In some, but not all, examples, the data collection configuration information also indicates the information that needs to be included in the data collection request.

[0080] In some, but not all, examples, the information that needs to be included in the data collection request includes one or more of the following: information associated with the UE’s supplier; information associated with the purpose of data collection; or an indication of the amount of data requested.

[0081] In some, but not necessarily all, examples, box 204 is an optional box that can be omitted from the method.

[0082] At box 208, the method includes receiving data collection configuration information from a network node. The data collection configuration information indicates the permission conditions for the data collection request.

[0083] At box 210, the method includes sending a request for data collection to a network node.

[0084] This request is based on the license terms for the indicated data collection request.

[0085] In some, but not all, examples, a request for data collection indicates a request for data associated with one or more UE-side ML models.

[0086] In some, but not all, examples, the requested data includes information associated with channel measurements and / or information associated with truth labels.

[0087] In some, but not all, examples, the information associated with channel measurements includes one or more of the following: channel measurement data; quality indicators associated with the channel measurement data; or timestamps associated with the channel measurement data. In some, but not all, examples, the information associated with channel measurements includes data suitable for unsupervised learning by ML models.

[0088] In some, but not all, examples, the information associated with the truth label includes one or more of the following: the truth label; an approximation of the truth label; quality indicators(s) associated with the truth label; or timestamps(s) associated with the truth label. In some, but not all, examples, the information associated with the truth label includes empirical or verifiable data suitable for supervised learning by an ML model.

[0089] In some, but not all, examples, requests for data collection are sent as part of one or more of the following: a Mobile Origin Location Request (MO-LR); an LPP request for auxiliary data; or a new LPP or Non-Access Stratum (NAS) protocol and / or message.

[0090] In some, but not all, examples, requests for data collection are sent as separate requests.

[0091] At box 214, the method includes receiving data from a network node associated with a sent request for data collection.

[0092] In some, but not all, examples, the received data corresponds to the requested data indicated in the request for data collection.

[0093] In some, but not all, examples, the received data includes data associated with channel measurements and / or information associated with truth labels.

[0094] At box 216, the method includes using the received data.

[0095] In some, but not all, examples, using the received data includes using the received data to train one or more machine learning models.

[0096] In some, but not all, examples, the use of received data includes using the received data to monitor one or more ML models.

[0097] In some, but not all, examples, using the received data includes using the received data to fine-tune one or more ML models in the ML model.

[0098] In some, but not all, examples involve using the received data to perform inference using one or more ML models and the received data.

[0099] In some, but not all, examples, using the received data includes sending the data to the UE-side node. In some, but not all, examples, the UE-side node to which the data is sent is configured to utilize one or more machine learning models. In some, but not all, examples, utilizing one or more machine learning models includes one or more of the following: using the received data to train one or more machine learning models; using the received data to monitor one or more machine learning models; using the received data to fine-tune one or more machine learning models; or using one or more machine learning models and the received data to perform inference.

[0100] In some, but not all, examples, the UE-side node is the UE-side server. In some, but not all, examples, the UE-side node is a node outside of network 100.

[0101] therefore, Figure 2 A method is shown, including: At box 202, it is determined whether one or more UE-side machine learning models require data collection; at box 204, a request for configuration information associated with a data collection request sent by the UE and associated with one or more UE-side machine learning models is sent to the network node; at box 208, the configuration information associated with the data collection request sent by the UE and associated with one or more UE-side machine learning models is received from the network node; at box 210, a request for data collection is sent to the network node; at box 214, data associated with the sent request for data collection is received from the network node; and at box 216, the received data is used.

[0102] From the perspective of the network node, the method includes: at block 204, receiving a request from the UE for data collection configuration information. At box 206, the method includes determining data collection configuration information.

[0103] In some, but not all, examples, data collection configuration information is determined for individual UEs. In some, but not all, examples, data collection configuration information is determined for all UEs served by the network node. In some, but not all, examples, data collection configuration information is determined for some, but not all, UEs served by the network node.

[0104] In some, but not all, examples, the determination of the data collection configuration is based on a trigger.

[0105] In some, but not all, examples, the trigger is receiving a request for data collection configuration information from the UE. In some such examples, the data collection configuration information is determined for the individual UE from which it receives the request for data collection configuration information.

[0106] In some, but not all, examples, the trigger is an internal trigger caused by a network node. In some such examples, data collection configuration information is determined for more than one UE served by a network node.

[0107] In some, but not all, examples, the data collection configuration information indicates one or more space-based constraints on the UE's data collection request. In some, but not all, examples, the data collection configuration information indicates one or more time-based constraints on the UE's data collection request.

[0108] By determining spatial and / or temporal constraints on UE data collection requests, network nodes proactively allocate data collection-related loads in both time and space. This facilitates load balancing across the network, such as preventing or adapting to large data flows in certain cells and / or time periods.

[0109] In some, but not all, examples, determining spatial and / or temporal constraints on UE data collection requests includes determining the data load for at least one location within at least one time period.

[0110] In some, but not all, examples, the data load is determined based on one or more of the following: the number of data collection requests received from the area; the frequency of data collection requests received from the area; information associated with reference signals configured in the area; information associated with active reference signal configurations in the area; the number of reference signal configurations determined and / or rejected by the gNB; the rate at which reference signal configurations are determined and / or rejected by the gNB; and the duration of reference signal configurations determined and / or rejected by the gNB.

[0111] In some, but not all, examples, spatial and / or temporal constraints for determining UE data collection requests also include determining the data load and / or relative data load for one or more network areas and / or one or more time periods based on the determined data load.

[0112] In some, but not all, examples, determining spatial and / or temporal constraints for UE data collection requests includes identifying one or more network areas and / or one or more time periods with the lowest relatively defined data load, and identifying the identified one or more network areas and / or one or more time periods as the network areas and / or time periods for which data collection requests are permitted.

[0113] In some, but not all, examples, determining spatial and / or temporal restrictions on UE data collection requests includes identifying one or more network areas and / or one or more time periods with the highest, relatively definite data load, and identifying the identified one or more network areas and / or one or more time periods as network areas and / or time periods where data collection requests are not permitted.

[0114] In some, but not all, examples, determining spatial and / or temporal restrictions on UE data collection requests includes identifying one or more network areas and / or one or more time periods with a determined data load below a threshold, and identifying the determined one or more network areas and / or one or more time periods as network areas and / or time periods permitted for data collection requests.

[0115] In some, but not all, examples, determining spatial and / or temporal restrictions on UE data collection requests includes identifying one or more network areas and / or one or more time periods with a determined data load exceeding a threshold, and identifying the identified one or more network areas and / or one or more time periods as network areas and / or time periods where data collection requests are not permitted.

[0116] In some, but not all, examples, the data collection configuration information indicates one or more configuration-based restrictions on the UE's data collection request.

[0117] In some, but not all, examples, configuration-based constraints for determining a UE data collection request include determining indications of one or more configurations used by the network node to collect data. In some, but not all, examples, configuration-based constraints for determining a UE data collection request include determining indications of the data available at the network node for transmission to the UE.

[0118] In some, but not all, examples, the indication of one or more configurations used by a network node to collect data indicates one or more parameters of reference signals and / or measurements used by the network node to collect data from one or more UEs.

[0119] In some, but not all, examples, the reference signals and / or one or more parameters used by the network node to collect data from one or more UEs depend at least on the network-side conditions during data collection. Examples of network-side conditions during data collection include: network data load; gNB capability; or gNB availability.

[0120] In some, but not all, examples, the reference signals and / or one or more parameters used by a network node to collect data from one or more UEs depend at least on the capabilities of the one or more UEs from which the data is collected. Example capabilities include support for one or more positioning reference signals (PRS) and / or measurement configurations.

[0121] By determining indications of one or more configurations used by the network node to collect data, the network node determines indications of the data already collected by the network node. This, in turn, provides indications of the availability of one or more types of data. Therefore, the configuration-based limitation of UE data collection requests is based on the data that the network node can use to transmit to the UE and / or the configurations used by the network node to collect data.

[0122] In some, but not all, examples, the data collection configuration information indicates one or more data volume-based restrictions on the UE's data collection request. In some, but not all, examples, the data collection configuration information indicates one or more purpose-based restrictions on the UE's data collection request.

[0123] By determining data volume-based limits for UE data collection requests, network nodes suppress data collection requests that they cannot fulfill, such as requests for a number of samples greater than the number of samples available to the network node, or requests for a number of samples that may cause data congestion of radio resources.

[0124] The purpose of a data collection request can indicate the required data volume and / or time requirements. For example, a data collection request for monitoring requires a small amount of data as quickly as possible, while a data collection request for training requires a large amount of data, but not immediately. Therefore, by determining the purpose-based constraints of UE data requests, network nodes proactively determine the data volume and timeliness requirements they will meet for data collection and reduce the number of data collection requests that will be rejected.

[0125] In some, but not all, examples, determining data collection requests based on data volume and / or purpose includes indicating the size of one or more datasets maintained by network nodes.

[0126] In some, but not all, examples, the indication for determining the size of one or more datasets maintained by a network node includes determining the size of one or more datasets stored in memory at the network node. In some, but not all, examples, the indication for determining the size of one or more datasets maintained by a network node includes determining the size of one or more datasets otherwise available at the network node.

[0127] In some, but not all, examples, an indication of the size of one or more datasets maintained by a network node indicates the total number of data samples that the network node can use for transmission.

[0128] In some, but not all, examples, the indication of the size of one or more datasets maintained by a network node includes: an indication of the size of a first dataset containing data of a first type, an indication of the size of a second dataset containing data of a second type, and an indication of the size of an nth dataset containing data of an nth type.

[0129] In some, but not all, examples, determining data collection requests based on data volume and / or purpose also includes determining the maximum amount of data available for the request based on an indication of the size of one or more datasets maintained by the network node.

[0130] In some, but not all, examples, the maximum amount of data available for a request is less than an indication of the size of one or more datasets maintained by the network node. In some, but not all, examples, the maximum amount of data available for a request is the same as an indication of the size of one or more datasets maintained by the network node.

[0131] In some, but not all, examples, the data collection configuration information indicates one or more device-based restrictions on the UE's data collection request. In some, but not all, examples, the data collection configuration information indicates one or more vendor-based restrictions on the UE's data collection request.

[0132] By determining device-based and / or vendor-based restrictions on UE data collection requests, network nodes prohibit excessive requests from a single device, from devices of the same type, or from devices with the same vendor.

[0133] In some, but not all, examples, determining device-based and / or vendor-based limitations of UE data collection requests includes identifying information associated with past data collection requests linked to different vendor IDs and / or different device IDs.

[0134] In some, but not all, examples, the information associated with past collection requests linked to a device ID includes one or more indications of the device ID and one or more of the following associated with the device ID: an indication of the number of data collection requests received in the past; or the rate of data collection requests received in the past.

[0135] In some, but not all, examples, the information associated with past collection requests linked to a vendor ID includes one or more indications of the vendor ID and one or more of the following associated with the vendor ID: an indication of the number of data collection requests received in the past; or the rate of data collection requests received in the past.

[0136] In some, but not all, examples, determining device-based and / or vendor-based limitations for UE data collection requests also includes: determining an indication of the number of past data collection requests and / or an indication of the relative number of past data collection requests associated with a device ID and / or a vendor ID, based on information identified as being associated with past collection requests from a device ID and / or a vendor ID.

[0137] In some, but not all, examples, determining device-based and / or vendor-based restrictions on UE data collection requests involves identifying one or more device IDs and / or one or more vendor IDs associated with a minimum relative number of past data collection requests, and identifying the identified one or more device IDs and / or one or more vendor IDs as device IDs and / or vendor IDs that allow data collection requests.

[0138] In some, but not all, examples, determining device-based and / or vendor-based restrictions on UE data collection requests includes identifying one or more device IDs and / or one or more vendor IDs associated with the highest relative number of past data collection requests, and identifying the identified one or more device IDs and / or one or more vendor IDs as device IDs and / or vendor IDs that are not permitted to make data collection requests.

[0139] In some, but not all, examples, determining device-based and / or vendor-based restrictions on UE data collection requests includes identifying one or more device IDs and / or one or more vendor IDs associated with a minimum number of past data collection requests, and identifying the identified one or more device IDs and / or one or more vendor IDs as device IDs and / or vendor IDs that allow the data collection requests.

[0140] In some, but not all, examples, determining device-based and / or vendor-based restrictions on UE data collection requests includes identifying one or more device IDs and / or one or more vendor IDs associated with the highest number of past data collection requests, and identifying the identified one or more device IDs and / or one or more vendor IDs as device IDs and / or vendor IDs that are not permitted to make data collection requests.

[0141] In some, but not all, examples, the data collection configuration information also indicates the information that needs to be included in the data collection request.

[0142] By indicating the information that needs to be included in a data collection request, network nodes can easily determine whether a data collection request complies with the permission conditions for that request.

[0143] At box 208, the method includes sending determined data collection configuration information to the UE.

[0144] In some, but not all, examples of determining data collection configuration information for a single UE, sending the determined data collection configuration information includes sending the determined data collection configuration information only to the UE that has determined the data collection configuration information.

[0145] In some, but not all, examples, where data collection configuration information is determined for more than one UE, sending the determined data collection configuration information includes sending the determined data collection configuration information to all UEs served by the network node. In some, but not all, examples, where data collection configuration information is determined for more than one UE, sending the determined data collection configuration information includes sending the determined data collection configuration information to fewer than all UEs served by the network node, including sending it to only one UE.

[0146] In box 210, the method includes receiving a request for data collection from the UE.

[0147] In box 212, the method includes determining data to be transmitted to the UE.

[0148] In some, but not all, examples, determining the data to be transmitted to the UE includes determining whether a received request for data collection complies with the permission conditions for the data collection request.

[0149] In some, but not all, examples, determining the data to be transmitted to the UE includes: rejecting the data collection request based on the determination that the received request for data collection does not meet the permission conditions for the data collection request.

[0150] In some, but not all, examples, determining the data to be transmitted to the UE includes initiating a process for collecting new data.

[0151] In block 214, the method includes transmitting the determined data to the UE.

[0152] therefore, Figure 2 A method is shown, including: At box 204, a request for configuration information associated with a data collection request is received from the UE. The data collection request is sent by the UE and associated with one or more UE-side machine learning models. At box 206, configuration information associated with a data collection request sent by the UE and associated with one or more UE-side machine learning models is determined; At box 208, configuration information associated with a data collection request sent by the UE and associated with one or more UE-side machine learning models is sent to the UE; at box 210, a request for data collection is received from the UE; at box 212, data to be sent to the UE is determined; and at box 214, the determined data is sent to the UE.

[0153] It should be understood that one or more boxes of method 200 may be omitted and / or rearranged. For example, in some, but not necessarily all, examples, box 204 may be optional. In some such examples, box 202 may be executed after box 208, and box 210 may be executed based on the execution of box 202.

[0154] Figure 3 An example of method 300 is shown.

[0155] Method 300 can be performed by any suitable means including any suitable components for performing the method (e.g., as per [reference]). Figure 6 The device described is used to perform this action.

[0156] In the example, method 300 can be executed by terminal node 110 (such as UE).

[0157] In box 302, the method includes determining whether one or more UE-side machine learning models require data collection.

[0158] At box 304, the method includes sending a request to the network node for data collection configuration information.

[0159] At box 306, the method includes receiving data collection configuration information from a network node.

[0160] At box 308, the method includes sending a request for data collection to a network node.

[0161] At box 310, the method includes receiving data from a network node associated with a sent request for data collection.

[0162] At box 312, the method includes using the received data.

[0163] therefore, Figure 3 A method is shown, including: At box 302, it is determined whether one or more UE-side machine learning models require data collection; at box 304, a request for configuration information associated with a data collection request sent by the UE and associated with one or more UE-side machine learning models is sent to the network node; at box 306, the configuration information associated with the data collection request sent by the UE and associated with one or more UE-side machine learning models is received from the network node; at box 308, a request for data collection is sent to the network node; at box 310, data associated with the sent request for data collection is received from the network node; and at box 312, the received data is used.

[0164] Figure 4A An example of method 400 is shown.

[0165] Method 400 can be performed by any suitable means including any suitable components for performing the method (e.g., as per [reference]). Figure 6 The device described is used to perform this action.

[0166] In the example, method 400 can be executed by terminal node 110 (such as UE).

[0167] At box 402, the method includes receiving data collection configuration information from a network node.

[0168] At box 404, the method includes sending a request for data collection to a network node based on indicated permission conditions for the data collection request.

[0169] therefore, Figure 4A A method is shown, including: At box 402, configuration information associated with a data collection request sent by the UE and associated with one or more UE-side machine learning models is received from the network node, wherein the configuration information indicates permission conditions for the data collection request; and At box 404, a request for data collection is sent to the network node based on the indicated permission conditions for the data collection request.

[0170] Figure 4B An example of method 450 is shown.

[0171] Method 450 can be performed by any suitable means including any suitable components for performing the method (e.g., as per [reference]). Figure 6The device described is used to perform this action.

[0172] In the example, method 450 can be executed by access node 120.

[0173] In some, but not all, examples, access node 120 is a network node that performs one or more network functions. In some, but not all, examples, the access node provides LMF.

[0174] In some, but not all, examples, access node 120 is a gNB.

[0175] At box 452, the method includes determining data collection configuration information.

[0176] At box 454, the method includes sending data collection configuration information to the UE.

[0177] At box 456, the method includes receiving a data collection request from the UE.

[0178] therefore, Figure 4B Method 450 is shown, including: At box 452, configuration information associated with a data collection request sent by a user equipment (UE) and associated with one or more UE-side machine learning models is determined, wherein the configuration information indicates permission conditions for the data collection request; at box 454, the configuration information is sent to the UE; and at box 456, the data collection request is received from the UE.

[0179] Figure 5 An example of method 500 is shown.

[0180] Method 500 can be performed by any suitable means including any suitable components for performing the method (e.g., as per [reference]). Figure 6 The device described is used to perform this action.

[0181] In the example, method 500 can be executed by access node 120.

[0182] In some, but not all, examples, access node 120 is a network node that performs one or more network functions. In some, but not all, examples, the access node provides LMF.

[0183] In some, but not all, examples, access node 120 is a gNB.

[0184] In box 502, the method includes receiving a request from the UE for data collection configuration information. At box 504, the method includes determining data collection configuration information.

[0185] At box 506, the method includes sending determined data collection configuration information to the UE.

[0186] In box 508, the method includes receiving a request for data collection from the UE.

[0187] In block 510, the method includes determining data to be transmitted to the UE.

[0188] In block 512, the method includes transmitting the determined data to the UE.

[0189] therefore, Figure 5 A method is shown, including: At box 502, a request for configuration information associated with a data collection request is received from the UE. The data collection request is sent by the UE and associated with one or more UE-side machine learning models. At box 504, configuration information associated with a data collection request sent by the UE and associated with one or more UE-side machine learning models is determined; At box 506, configuration information associated with a data collection request sent by the UE and associated with one or more UE-side machine learning models is sent to the UE; at box 508, a request for data collection is received from the UE; at box 510, data to be sent to the UE is determined; and at box 512, the determined data is sent to the UE.

[0190] To collect the necessary data for the UE-side AI / ML model used for positioning, the UE can request it from the NW (e.g., LMF): Location-related channel measurements along with quality indicators and timestamps, and / or The associated truth label or its approximation, along with the quality indicator and timestamp.

[0191] However, UE requests for data collection may pose challenges to the NW (network), including the following: The UE can send requests to the NW at any time, which can consume a lot of radio resources if it happens too frequently, and the NW may not even process and respond to such a number of requests, for example, depending on the positioning / processing load at the LMF.

[0192] Similarly, a vendor may instruct all its UEs to send the same or similar requests to the network simultaneously and repeatedly, resulting in too many redundant requests consuming radio resources.

[0193] The UE may send overly demanding requests, such as large amounts of partial A and / or partial B data, which the network may not be able to fulfill depending on data availability and resources.

[0194] Therefore, an effective mechanism for initiating data collection of AI / ML models on the network control UE side becomes necessary.

[0195] This disclosure provides a method for a network (NW) to control a UE to request data collection for a UE-side AI / ML model used for positioning.

[0196] This disclosure discloses steps for the NW to provide the UE with configuration for controlling the UE's requests for data collection (i.e., for requesting data from the LMF) via one or more of the following: Region-based and / or time-based restrictions: These allow the UE to request data only in certain network areas and only during certain time frames. This is to allow the UE to send any request to the NW to request data, wherever and when. The NW may not know in advance where the DC request might occur (e.g., certain cells serving the UE). Using these configurations, the NW can proactively allocate DC-related loads across time and space.

[0197] Configuration or network-side condition-based restrictions: The UE is allowed to request data associated with certain configurations or network-side conditions, such as data associated with certain configurations / network-side conditions for which the network identifier (associated ID) is known.

[0198] Based on data volume limitations: UEs are only permitted to request data collection associated with a specific data volume, such as a maximum number of data samples. The data volume is primarily related to the content requested by the data UE; for example, a very large number of samples that the NW may not be able to provide in time upon request, or that the data volume may be unable to be transmitted due to potential congestion on radio resources.

[0199] Purpose-based restrictions: The UE is only allowed to request data associated with certain purposes (e.g., training, monitoring). Note that the purpose indirectly relates to the UE's data requirements, such as data volume and timeliness. For example, for monitoring, the UE desires a relatively small amount of data but as quickly as possible, while for training, the UE desires a much larger dataset, not immediately. Therefore, the NW can inform it in advance which types of requests are allowed (and where and when, as mentioned above).

[0200] Device-based restrictions: Data requests are only allowed from certain UEs / chipsets of certain vendors, or subject to the above restrictions. This is to avoid any vendor-specific spam from NW, i.e., a large / frequent number of requests from the same vendor.

[0201] Return to reference Figure 2 The following method box is shown: 202: The UE determines that it needs to collect data from the NW, for example, to train an AI / ML model for localization purposes.

[0202] 204: The UE can control the configuration of the UE's data collection (DC) request from the NW request. Note that this step is optional.

[0203] The motivation for the UE to obtain configuration in advance is to know the DC limits imposed by the network before sending any requests, thus avoiding the risk of requests being rejected by the NW. This allows the UE to avoid unnecessarily expending resources (i.e., avoiding sending requests that may be rejected by the NW).

[0204] 206: The NW (e.g., LMF) determines the configuration used to control UE request data collection (DC), which may contain at least one or more of the following information: Spatial-based restrictions on UE DC requests: These specify certain areas, determined by the NW, where UE data requests are (not) permitted, such as geographical areas, tracking areas, RAN notification areas (RNA), serving cell / gNB, and cell / gNB / TRP lists. Time-based restrictions on UE DC requests: These specify a particular time, determined by the NW, during which the UE is (not) allowed to request data, for example, according to a time window.

[0205] These limitations enable the NW to distribute the load associated with data collection spatially and temporally to achieve load balancing across the network. For example, a network may need to serve a large volume of data traffic in certain cells, and it may not be able to handle a large number of requests from UEs serving in those cells during certain time periods. The network can even identify dedicated cells that allow UEs to send their requests for data collection purposes.

[0206] To determine the above configuration, the LMF can first identify the load at different locations in its service area based on at least one or more of the following: the number / frequency of location requests associated with UEs served by different cells, information related to the configuration / active PRS transmissions in different cells, and the number / rate / duration of PRS configurations determined / rejected by different cells (configured by the gNB based on requests from the LMF. If the gNB rejects the configuration, it can indicate high load in that gNB). Furthermore, the LMF can identify cells with lower load and create a list of these cells.

[0207] Configuration-based restrictions: These specify the range of data collection-related configurations that are (not) permitted for UE requests, such as the maximum number of configurations, the range of associated IDs for determining PRS configurations, and the parameters of the PRS configurations used for data collection. Data volume limitations: Specifies the amount of data that the UE is allowed (or not allowed) to request, for example, based on the maximum number of samples.

[0208] Purpose-based restrictions: Specify the purpose for which the UE is not allowed to request data, such as (re)training, monitoring, or fine-tuning.

[0209] By leveraging the aforementioned limitations, the NW can proactively avoid unnecessary requests that it would otherwise reject regardless of the amount and purpose of data collection. For example, the size of the dataset at the network may be significantly limited compared to what the UE might need, or the NW can use data volume information when billing users for data collection. Similarly, the purpose of data collection can indicate the amount of data required; that is, for monitoring or retraining / updating an AI / ML model, the UE needs far less data than it would need to train from scratch.

[0210] To determine these limitations, the LMF can maintain a dataset consisting of measurement samples and / or associated truth labels, collected from PRUs served by the LMF. Based on the size of the dataset (e.g., the number of samples), the LMF can determine the maximum amount of data a UE can request, which is less than the size of its maintained dataset. Device-based limitations: Devices that are (not) permitted to send DC requests are specified, particularly for cases where the limitation is broadcast to all UEs. In embodiments, this refers to devices associated with certain vendors (e.g., UE and / or chipset vendors) that are (not) permitted to send DC requests.

[0211] This option allows NW to restrict or eliminate requests from certain vendors. For example, vendor X might want to greedily collect data and "attack" the network by ordering all its UEs to request data / PRS from the network, which could easily overload radio resources.

[0212] To determine the above, the LMF can maintain a list of vendor IDs (e.g., obtained from the AMF, or via a request from the UE as described below) and statistics / numbers of past requests related to location and / or data collection from each vendor. Based on this historical data, the LMF identifies the vendor that sends the most requests and limits its future requests.

[0213] The above restrictions are valid for a specific time / area. This configuration can also indicate the mandatory fields that the UE needs to specify when sending a request, such as: UE / chipset supplier information Purpose of data collection • The amount of data requested, for example, based on the number of samples. Note that box 206 can be triggered based on a configuration request from the UE or triggered internally by the LMF. In the first case, the LMF can determine a dedicated configuration for each UE individually, which can be provided to the UE via dedicated signaling. However, in the latter case, the LMF can determine a common configuration for all UEs, for example, which will be broadcast by one or more cells.

[0214] 208: NW provides the determined configuration to the UE.

[0215] 210: The UE sends its request for data based on the configuration received in the previous steps. For example, the UE sends the request within the permitted area and time frame, and by indicating mandatory information for the NW request in its request.

[0216] In embodiments, a request may be sent as part of MO-LR, LPP request auxiliary data, or another / new LPP or NAS protocol and / or message (e.g., LPP request auxiliary data for data collection), or it may be a standalone request (e.g., a request sent only for collecting measurements and not for positioning).

[0217] 212: Upon receiving a request, the NW determines the data to be delivered to the UE. To this end, if necessary, the NW may initiate a process of collecting new data, for example, by locating other UEs / PRUs.

[0218] 214: The NW / LMF provides the data it collects (e.g., truth information collected) to the UE. 216: UE uses the collected data to, for example, (re)train its AI / ML model.

[0219] In some embodiments, instead of LMF, another network entity (e.g., AMF, gNB, OAM) can perform the above roles.

[0220] The signaling between the NW and UE used in the above process can occur via different protocols, such as LPP (between LMF and UE), RRC (between gNB and UE), MAC (between gNB and UE), etc. The accompanying figures illustrate methods performed by a system involving interactions between different system entities. The figures also illustrate a collection of individual methods performed independently by the different system entities.

[0221] Figure 6 An example of a controller 600 suitable for use in device 610 is shown. The controller 600 can be implemented as a controller circuit. The controller 600 can be implemented solely in hardware, have certain aspects only in software including firmware, or can be a combination of hardware and software (including firmware).

[0222] like Figure 6 As shown, the controller 600 can be implemented using instructions that implement hardware functions, for example, by using executable instructions 606 in a general-purpose or special-purpose processor 602, which can be stored on a machine-readable storage medium (disk, memory, etc.) for execution by such processor 602.

[0223] Processor 602 is configured to read from and write to memory 604. Processor 602 may also include an output interface and an input interface, through which processor 602 outputs data and / or commands and through which data and / or commands are input to processor 602.

[0224] Memory 604 stores instructions, programs, or code 606 that control the operation of device 610 when loaded into processor 602. Instructions, programs, or code 606 provide logic and routines that enable device 610 to perform the methods shown in the figures.

[0225] The apparatus 610 includes: at least one processor 602; and at least one memory 604 storing instructions that, when executed by the at least one processor 602, cause the apparatus to at least: receive configuration information associated with a data collection request from a network node, the data collection request being sent by a UE and associated with one or more UE-side machine learning models, wherein the configuration information indicates permission conditions for the data collection request; and send a data collection request to the network node based on the indicated permission conditions for the data collection request.

[0226] In some examples, there exists a (computer-implemented) system that includes: A component for receiving configuration information associated with a data collection request from a network node, the data collection request being sent by a UE and associated with one or more UE-side machine learning models, wherein the configuration information indicates permission conditions for the data collection request; and a component for sending a data collection request to a network node based on the indicated permission conditions for the data collection request.

[0227] like Figure 7 As shown, instructions, programs, or code 606 can reach device 610 via any suitable delivery mechanism 608. Delivery mechanism 608 can be, for example, a machine-readable medium, a computer-readable medium, a non-transitory computer-readable storage medium, a computer program product, a memory device, a recording medium such as solid-state memory, or an article of manufacture that includes or tangibly embodies instructions 606. The delivery mechanism can be a signal configured to reliably transmit instructions 606. Device 610 can propagate or transmit instructions 606 as a data signal.

[0228] As used herein, the term “non-transitory” refers to a limitation on the medium itself (i.e., tangible, not tactile), rather than a limitation on the persistence of data storage (e.g., RAM versus ROM).

[0229] Instruction 606 causes the apparatus to perform at least the following operations: receive configuration information from a network node associated with a data collection request sent by a UE and associated with one or more UE-side machine learning models, wherein the configuration information indicates permission conditions for the data collection request; and send a data collection request to the network node based on the indicated permission conditions for the data collection request.

[0230] Instruction 606 may be included in a computer program, a non-transitory computer-readable medium, a computer program product, or a machine-readable medium. In some, but not all, examples, instruction 606 may be distributed across more than one computer program.

[0231] Although memory 604 is shown as a single component / circuit, it can be implemented as one or more separate components / circuits, some or all of which may be integrated / removable and / or provide permanent / semi-permanent / dynamic / cached storage.

[0232] Although processor 602 is shown as a single component / circuit, it can be implemented as one or more separate components / circuits, some or all of which may be integrated / removable. Processor 602 can be a single-core or multi-core processor.

[0233] References to “computer-readable storage medium,” “computer program product,” “tangible computer program,” or “controller,” “computer,” “processor,” etc., should be understood to encompass not only computers with different architectures (such as single / multiprocessor architectures and sequential (von Neumann) / parallel architectures) but also special-purpose circuits, such as field-programmable gate arrays (FPGAs), special-purpose circuits (ASICs), signal processing devices, and other processing circuits, including quantum processing circuits. References to computer programs, instructions, code, etc., should be understood to encompass software or firmware for programmable processors, such as the programmable content of hardware devices, whether instructions for processors or configuration settings for fixed-function devices, gate arrays, or programmable logic devices, etc. As used in this application, the term "circuit" may refer to one or more or all of the following: (a) Pure hardware circuit implementations (such as implementations in analog, digital, and / or quantum circuits) and (b) A combination of hardware circuitry and software, such as (if applicable): i. A combination of analog, digital, and / or quantum hardware circuitry with software / firmware, and ii. Any or all parts of a hardware processor (including (multiple) digital and / or quantum processors), together with software and (multiple) memories, that work together to enable a device, such as a mobile device, computing device, or server, to perform various functions, and (c) Any or all parts of the hardware circuitry, such as (multiple) microprocessors and / or quantum processors, that require software (e.g., firmware) to operate, but which may not exist when no software is required to operate.

[0234] This definition of "circuit" applies to all uses of the term in this application (including in any claim). As another example, as used herein, the term "circuit" also encompasses only hardware circuitry or a processor (or multiple processors) or a portion thereof and its accompanying software and / or firmware implementation. The term "circuit" also encompasses, for example and if applicable to elements of the claims, baseband integrated circuits or processor integrated circuits for mobile devices or similar integrated circuits in servers, cellular network devices, or other computing or network devices.

[0235] The boxes shown in the accompanying drawings may represent steps in the method and / or code segments in instruction 606. The description of a specific order of boxes does not necessarily imply a required or preferred order, and the order and arrangement of boxes can vary. Furthermore, some boxes may be omitted.

[0236] As used herein, a module refers to a unit or device that excludes certain parts / components that will be added by the final manufacturer or user. Device 610 may be, for example, a module. The controller 600 of device 610 may be, for example, a module.

[0237] Where a structural feature has been described, it may be replaced by means of one or more functions for performing the structural feature, whether the function or those functions are explicitly or implicitly described.

[0238] In some, but not all, examples, device 610 is configured to perform or not perform local storage of data in memory 604 at device 610 and to transfer data from device 610 with or without local processing of data by circuitry or processor at device 610.

[0239] The data may be, for example, measurement data from one or more terminal nodes and / or one or more network nodes, or data generated by processing measurement data from one or more terminal nodes and / or one or more network nodes.

[0240] Data can be stored remotely on one or more devices in processed or unprocessed formats. Data can also be stored in the cloud.

[0241] Data can be processed remotely at one or more devices. Data can be processed partly locally and partly remotely at one or more devices.

[0242] For example, data can be wirelessly transmitted to a remote device via short-range radio communication such as Wi-Fi or Bluetooth, or via a long-range cellular radio link. The device may include a communication interface, such as a radio transceiver for data communication.

[0243] Device 610 can be part of an Internet of Things (IoT) that forms part of a larger distributed network.

[0244] Data processing, whether local or remote, can be used for health monitoring, data aggregation, patient monitoring, vital sign monitoring, or other purposes.

[0245] Data processing, whether local or remote, can involve artificial intelligence or machine learning algorithms. Data can be used, for example, as learning input to train a machine learning network, or as query input to a machine learning network that provides a response. Machine learning networks can, for example, use linear regression, logistic regression, vector support machines, or non-recurrent machine learning networks such as neural networks with one or more hidden layers.

[0246] Data processing, whether local or remote, can generate output. This output can be transmitted to device 610, where it can produce output sensitive to the subject, such as audio output, visual output, or tactile output.

[0247] Systems, apparatuses, methods, and computer programs can use machine learning, which may include statistical learning. Machine learning is a field of computer science that gives computers the ability to learn without explicit programming. If a computer's performance (as measured by P) at a task in T improves with experience E, then the computer learns from experience E regarding a class of tasks T and the performance metric P. Computers can typically learn from previous training data to predict future data. Machine learning includes fully or partially supervised learning and fully or partially unsupervised learning. It can achieve discrete outputs (e.g., classification, clustering) and continuous outputs (e.g., regression). Machine learning can be implemented using, for example, different methods such as cost function minimization, artificial neural networks, support vector machines, and Bayesian networks. Cost function minimization can be used, for example, for linear and multinomial regression and K-means clustering. Artificial neural networks, for example, with one or more hidden layers, model complex relationships between input and output vectors. Support vector machines can be used for supervised learning. A Bayesian network is a directed acyclic graph representing the conditional independence of multiple random variables.

[0248] The above-described examples are particularly suitable for implementations where the design is driven by technical considerations of the system's or network's internal functions. These examples are designed to leverage specific technical properties of the technical systems on which they are implemented to achieve technical effects, such as efficient use of computer storage capacity, network bandwidth, and power consumption.

[0249] This method also assigns the execution of data-intensive training of machine learning algorithms to the client and the preparation steps to the server to take advantage of a server-client architecture.

[0250] Simplifying the training data and training of machine learning models is technically feasible because there is distributed training across multiple clients, and the training data at each client is secure and kept private.

[0251] Machine learning models can be applied to many technical fields. For example: - Classify digital images, videos, audio, or speech signals based on low-level features (e.g., image edges or pixel attributes).

[0252] - Control technology systems or processes, such as computer-controlled classification systems or industrial processes. - Determine the adaptation to the industrial process based on measurements; - Digital audio, image, or video enhancement or analysis -Source separation in speech signals; speech recognition, - Encoding data for reliable and / or efficient transmission or storage (and corresponding decoding); compression of audio, image, video, or sensor data; - Encrypt / decrypt or sign electronic communications; - Technical parameters (e.g., energy consumption, core temperature) are determined by processing data obtained from sensors; - Provide a reliability estimate of technical information (e.g., genotype). - Provide medical diagnostics through automated systems that process physiological measurements.

[0253] - Derive or predict the physical state of an existing real object from the measurement of its physical properties. - The sensor data provided as input to the ML model is causally linked to the control command output for the control device provided as output to the ML model.

[0254] The above examples can be applied as implementation components to: automotive systems; telecommunications systems; electronic systems including consumer electronics; distributed computing systems; media systems for generating or rendering media content including audio, visual, and audiovisual content, as well as mixed, mediated, virtual, and / or augmented reality media content; personal systems including personal health systems or personal fitness systems; navigation systems; user interfaces, also known as human-machine interfaces; networks including cellular, non-cellular, and optical networks; self-organizing networks; the Internet; the Internet of Things; virtualized networks; and related software and services.

[0255] According to the examples of this disclosure, the device can be provided in an electronic device (e.g., a mobile terminal). However, it should be understood that the mobile terminal is merely an illustration of an electronic device that will benefit from implementations of this disclosure and should therefore not be construed as limiting the scope of this disclosure. While the device may be provided in a mobile terminal in some implementation examples, other types of electronic devices, such as, but not limited to, mobile communication devices, handheld portable electronic devices, wearable computing devices, portable digital assistants (PDAs), pagers, mobile computers, desktop computers, televisions, gaming devices, laptop computers, cameras, video recorders, GPS devices, and other types of electronic systems, can readily adopt the examples of this disclosure. Furthermore, devices can readily adopt the examples of this disclosure regardless of their intention to provide mobility.

[0256] The term “includes” as used in this document has an inclusive rather than exclusive meaning. That is, any reference to X that includes Y indicates that X may include only one Y or may include more than one Y. If the intention is to use “includes” with an exclusive meaning, it will become clear in the context by referring to “includes only one…” or by using “consisting of…”.

[0257] In this specification, the terms “connection,” “coupling,” and “communication,” and their derivatives, mean operatively connecting / coupling / communicating. It should be understood that any number or combination of intermediate components (including none) may be present to provide direct or indirect connection / coupling / communication. Any such intermediate component may include hardware and / or software components.

[0258] As used herein, the term "determine / determine" (and its grammatical variations) can include at least: calculation, operation, processing, derivation, measurement, investigation, identification, lookup (e.g., searching in a table, database, or other data structure), ascertainment, etc. Furthermore, "determine" can include receiving (e.g., receiving information), accessing (e.g., accessing data in memory), obtaining, etc. Additionally, "determine / determine" can include parsing, selecting, picking, establishing, etc.

[0259] Various examples have been referenced in this specification. Features or functions described in connection with an example indicate that those features or functions exist in that example. The use of the terms "example," "for example," "can," or "may" in the text indicates that, whether explicitly stated or not, these features or functions exist at least in the described example, and whether or not they are described as examples, they may, but not necessarily, exist in some or all other examples. Therefore, "example," "for example," "can," or "may" refers to a specific instance of a class of examples. An instance's property can be a property of only that instance, or a property of the class, or a property of a subclass of a class that includes some, but not all, instances of that class. Therefore, it is implicitly disclosed that features described with reference to one example, and not another, may be used as part of a feasible combination in that other example where possible, but are not necessarily required to be used in that other example.

[0260] As used herein, “at least one of the following” and “at least one” and similar wording, where a list of two or more elements connected by “and” or “or” means at least any one of these elements, or at least any two or more elements, or at least all of the elements.

[0261] Although examples have been described with reference to various examples in the preceding paragraphs, it should be understood that modifications may be made to the given examples without departing from the scope of the claims.

[0262] The features described above can be used in combinations other than those explicitly described above.

[0263] Although some features have been described with reference to certain characteristics, these functions can be performed by other features, whether or not they are described.

[0264] A description of a feature (such as a device or a component of a device) configured to perform a function or for performing a function should also be considered as disclosing a method for performing that function. For example, a description of a device configured to perform one or more actions or for performing one or more actions should also be considered as disclosing a method for performing those one or more actions with or without the device.

[0265] Although features have been described with reference to some examples, these features may also exist in other examples, whether or not they are described.

[0266] The terms “a,” “an,” or “the” as used in this document are inclusive rather than exclusive. That is, any reference to X including a / an / the Y means that X may include only one Y or may include more than one Y, unless the context clearly indicates otherwise. If “a,” “an,” or “the” is intended to be used in an exclusive sense, it will be clearly stated in the context. In some cases, the use of “at least one” or “one or more” may be used to emphasize an inclusive meaning, but the absence of these terms should not be presumed to have any exclusive meaning.

[0267] The presence of a feature (or combination of features) in a claim is a reference to that feature or combination of features itself, as well as to features that achieve substantially the same technical effect (equivalent features). Equivalent features include, for example, features that are variations and achieve substantially the same result in substantially the same manner. Equivalent features include, for example, features that perform substantially the same function in substantially the same manner to achieve substantially the same result.

[0268] In this specification, various examples have been referenced to describe the characteristics of the examples using adjectives or adjective phrases. Such descriptions of the characteristics associated with the examples indicate that the characteristics exist exactly as described in some examples, and substantially as described in others.

[0269] As used herein, the terms “at least one” and “one or more” mean “any one of at least one” and “any one of one or more”, respectively.

[0270] The foregoing description illustrates some examples of this disclosure; however, those skilled in the art will recognize possible alternative structures and methodological features that provide equivalent functionality to the specific examples of such structures and features described above, and for the sake of brevity and clarity, have been omitted from the foregoing description. Nevertheless, the foregoing description should be understood to implicitly include references to such alternative structures and methodological features that provide equivalent functionality, unless such alternative structures or methodological features are explicitly excluded in the foregoing description of the examples of this disclosure.

[0271] Although efforts have been made in the foregoing specification to draw attention to those features deemed important, the applicant may seek protection by means of the claims for any patentable feature or combination of features shown in the foregoing references and / or the accompanying drawings, whether or not they have been emphasized.

[0272] The following are some example implementations of embodiments according to this disclosure, which may be combined in any suitable manner.

[0273] Example 1. A user equipment (UE) for communication, comprising: At least one processor and at least one memory, the at least one memory storing instructions that, when executed by the at least one processor, cause the UE to execute at least the following: Receive configuration information associated with a data collection request from a network node, the data collection request being sent by the UE and associated with one or more UE-side machine learning models, wherein the configuration information indicates permission conditions for the data collection request; and Based on the indicated permission conditions for the data collection request, a data collection request is sent to the network node.

[0274] Example 2. Based on the user equipment of Example 1, the received configuration information indicates one or more of the following: One or more network areas that are authorized to collect data; Permission to collect data for one or more time periods; Configuration of one or more reference signals for the permission data request; One or more measurement and reporting configurations for the permission data request; The maximum amount of data that can be requested; One or more permission purposes for requesting data; One or more device types for which permission data collection is requested; or One or more vendors that grant permission to collect data.

[0275] Example 3. Based on the user equipment in Example 1, where the configuration information also indicates the information that needs to be included in the data collection request.

[0276] Example 4. According to the user equipment of Example 3, the information that needs to be included in the data collection request includes one or more of the following: information associated with the UE's supplier; information associated with the purpose of data collection; or an indication of the amount of data requested.

[0277] Example 5. A user equipment according to any one of Examples 1 to 4, wherein the instruction, when executed by at least one processor, also causes the UE to perform: Determine whether one or more machine learning models require data collection; and Based on the determination that one or more machine learning models require data collection, a request for data collection is sent to network nodes based on configuration information.

[0278] Example 6. A user equipment according to any one of Examples 1 to 4, wherein the instruction, when executed by at least one processor, also causes the UE to perform: Determine whether one or more machine learning models require data collection; and Based on the determination that one or more machine learning models require data collection, a request is sent to the network node for configuration information associated with the data collection request sent by the UE and associated with one or more UE-side machine learning models.

[0279] Example 7. A user equipment according to any one of Examples 1 to 4, wherein the instruction, when executed by at least one processor, also causes the UE to perform: Receive data from a network node associated with a data collection request; and one or more of the following: Use the received data to train one or more machine learning models in the machine learning model; Use the received data to monitor one or more machine learning models in the machine learning model; Use the received data to fine-tune one or more machine learning models in the machine learning model; Inference is performed using one or more machine learning models and the received data.

[0280] Example 8. A user equipment according to any one of Examples 1 to 4, wherein the data collection request includes a request for collecting data associated with one or more of the following: channel measurements; or truth labels.

[0281] Example 9. A user equipment according to any one of Examples 1 to 4, wherein the instructions, when executed by at least one processor, also cause the UE to perform: using one or more machine learning models from the machine learning models for one or more of the following: positioning; beam management; mobility; resource allocation; data transmission; or data reception.

[0282] Example 10. A method for communication, comprising: Receive configuration information associated with a data collection request from a network node, the data collection request being sent by the UE and associated with one or more UE-side machine learning models, wherein the configuration information indicates permission conditions for the data collection request; and Based on the indicated permission conditions for the data collection request, a data collection request is sent to the network node.

[0283] Example 11. A network node for communication, comprising: At least one processor and at least one memory, the at least one memory storing instructions that, when executed by the at least one processor, cause the network node to execute at least the following: Determine the configuration information associated with the data collection request, which is sent by the user equipment (UE) and associated with one or more UE-side machine learning models, wherein the configuration information indicates the permission conditions for the data collection request; Send configuration information to the UE; and Receive data collection requests from the UE.

[0284] Example 12. Based on the network node in Example 11, the configuration information indicates one or more of the following: One or more network areas that are authorized to collect data; Permission to collect data for one or more time periods; Configuration of one or more reference signals for the permission data request; One or more measurement and reporting configurations for the permission data request; The maximum amount of data that can be requested; One or more permission purposes for requesting data; One or more device types for which permission data collection is requested; or One or more vendors that grant permission to collect data.

[0285] Example 13. Based on the network nodes of Example 11 or Example 12, the configuration information includes one or more of the following: Determine the data load at at least one location within at least one time period, and determine one or more network areas and / or one or more time periods for the permitted data collection request based on the determined data load; Determine the size of the dataset maintained by the network nodes, and based on the determined size of the dataset, determine the maximum amount of data that can be requested; Determine the number of past data collection requests received from at least one device and / or at least one vendor, and determine one or more devices and / or one or more device vendors that authorize the data collection requests based on the number of past data collection requests received.

[0286] Example 14. A network node according to Example 11 or Example 12, wherein the instructions, when executed by at least one processor, also cause the network node to perform: Based on the received data collection request, determine the data to be sent to the UE; and Send the determined data to the UE.

[0287] Example 15. A method for communication, comprising: The configuration information associated with the data collection request is determined; the data collection request is sent by the user equipment (UE) and associated with one or more UE-side machine learning models, wherein the configuration information indicates the permission conditions for the data collection request; Send configuration information to the UE; and Receive a request for data collection from the UE.

Claims

1. A user equipment (UE) for communication, comprising: At least one processor and at least one memory, the at least one memory storing instructions that, when executed by the at least one processor, cause the UE to perform at least the following: Receive configuration information associated with a data collection request sent by the UE and associated with one or more UE-side machine learning models from a network node, wherein the configuration information indicates permission conditions for the data collection request; as well as The request for data collection is sent to the network node based on the indicated permission conditions for the data collection request.

2. The user equipment according to claim 1, wherein the received configuration information indicates one or more of the following: One or more network areas are permitted to make the data collection request; Permission is granted for one or more time periods for the data collection request; Configuration of one or more reference signals for the permission data request; One or more measurement and reporting configurations for the permission data request; The maximum amount of data that can be requested; One or more permission purposes for requesting data; Permit one or more device types for the data collection request; or One or more providers may authorize the data collection request.

3. The user equipment of claim 1, wherein the configuration information further indicates information that needs to be included in the data collection request.

4. The user equipment of claim 3, wherein the information to be included in the data collection request includes one or more of the following: information associated with the UE's supplier; information associated with the purpose of the data collection; or an indication of the requested amount of data.

5. The user equipment according to any one of claims 1 to 4, wherein the instructions, when executed by the at least one processor, further cause the UE to perform: Determine whether the one or more machine learning models require data collection; and Based on the determination that one or more machine learning models require data collection, a request for data collection is sent to the network node based on the configuration information.

6. The user equipment according to any one of claims 1 to 4, wherein the instructions, when executed by the at least one processor, further cause the UE to perform: Determine whether the one or more machine learning models require data collection; and Based on the determination that the one or more machine learning models require data collection, a request is sent to the network node for configuration information associated with the data collection request sent by the UE and associated with the one or more UE-side machine learning models.

7. The user equipment according to any one of claims 1 to 4, wherein the instructions, when executed by the at least one processor, further cause the UE to perform: Receive data associated with the data collection request sent from the network node; And one or more of the following: Use the received data to train one or more machine learning models in the machine learning model; Use the received data to monitor one or more machine learning models in the machine learning model; Use the received data to fine-tune one or more machine learning models in the machine learning model; Inference is performed using one or more of the machine learning models and the received data.

8. The user equipment according to any one of claims 1 to 4, wherein the data collection request includes a request for collecting data associated with one or more of the following: channel measurements; or truth labels.

9. The user equipment according to any one of claims 1 to 4, wherein the instructions, when executed by the at least one processor, further cause the UE to perform: using one or more of the machine learning models in the machine learning models for one or more of the following: positioning; beam management; mobility; resource allocation; data transmission; or data reception.

10. A method for communication, comprising: Receive configuration information associated with a data collection request sent by the UE and associated with one or more UE-side machine learning models from a network node, wherein the configuration information indicates permission conditions for the data collection request; as well as The request for data collection is sent to the network node based on the indicated permission conditions for the data collection request.

Images