Method for identifying valid data samples for ai / ml model training for positioning

Abstract

Various embodiments provide a method for determining whether positioning data should be collected and used as training data for training artificial intelligence / machine learning (AI / ML) models. A signaling mechanism can be provided in which a network node specifies which positioning measurement data should be collected for training purposes and which data should be discarded by providing criteria for data collection to a UE or gNB. A first network node can determine a tolerance for ground truth labeling (512), and then based on the tolerance, determine one or more reliability criteria for positioning data (514). The first network node can provide a request to a second network node for positioning data and reliability criteria (516), the second network node can collect positioning data based on the criteria (518), and then provide the collected positioning data to the first network node (522), which can then use the positioning data as training data.

Description

[0001] Related applications

[0002] This application claims the benefit of provisional patent application serial number 63 / 570,445, filed on March 27, 2024, the entire contents of which are hereby universally referenced and incorporated herein by reference. Technical Field

[0003] This disclosure relates to methods for determining whether location data should be collected and used as training data to train artificial intelligence / machine learning (AI / ML) models in wireless communication systems. Background Technology

[0004] Artificial intelligence (AI) or machine learning (ML) techniques consist of one or more algorithms that use a set of data as input to train one or more AI / ML models. The output of this AI / ML model is used by a device (e.g., a user equipment (UE), base station (BS), or another node) to perform certain actions or make certain decisions (e.g., handover) based entirely or partially on predictions, which in turn depend on the trained model. This AI / ML model can be trained online on the device (or in real time while processing data) or offline in the background. More specifically: Online training is an AI / ML training process in which the model used for inference is trained (near) in real time as new training samples or data arrive. Offline training is an AI / ML training process in which a model is trained based on collected samples or data, and the trained model is then used or delivered for inference.

[0005] AI / ML model inference refers to the process of using a trained AI / ML model to generate a set of outputs based on a set of inputs.

[0006] AL / ML models can be trained on a device, which can be a UE, a network node, or another node. In this respect, AL / ML models can be broadly classified as: Example 1: UE-side (AI / ML) model. This is an AI / ML model where inference is performed entirely at the UE. Example 2: Network-side (AI / ML) model. This is an AI / ML model where inference is performed entirely at the network level. Example 3: One-sided (AI / ML) model. This can be either a UE-side (AI / ML) model or a network-side (AI / ML) model. Example 4: Two-sided (AI / ML) model. It is a paired AI / ML model on which joint inference is performed, where joint inference includes AI / ML inference, which is jointly performed on the UE and the network. That is, the first part of the inference is first performed by the UE, and then the remaining part is performed by the gNB, or vice versa.

[0007] AI / ML models can be transmitted or delivered via an air interface based on one or more parameters of the known model structure at the receiving end, or a new model with parameters. Model delivery can include a complete model or a partial model.

[0008] The term AI / ML model lifecycle management (LCM) refers to the process of developing, deploying, and maintaining AI / ML models. The AI / ML model training pipeline includes several processing phases, such as collecting unprocessed input data from a data repository (data ingestion), finding high-quality input features (data preprocessing), finding the optimal mapping from the model input features to the desired model output target in the sense determined by the loss function (model training), and evaluating the model's performance on unseen data at the functional level and, when relevant, at the system level (model evaluation). The training pipeline typically concludes with a model registration phase, which may include operations such as compiling the ML model to specific hardware to make it runnable, and steps such as versioning and packaging the model so that it can be executed. Figure 1 The document shows examples illustrating different stages of an AI / ML model training pipeline.

[0009] According to the Technical Specification Group for Radio Access Networks; Research on Artificial Intelligence (AI) / Machine Learning (ML) for the NR Air Interface (Revision 18), 3GPP TR 38.843, December 2023, LCM covers the following components: Data collection Model training Function / Model Recognition Model Inference Function / model selection, activation, deactivation, switching, and rollback operations Function / Model Monitoring Model update UE capabilities.

[0010] In function-based LCM, the network indicates the activation / deactivation / fallback / switching of AI / ML functions via 3GPP signaling (e.g., Radio Resource Control (RRC), Media Access Control (MAC) Control Element (MAC-CE), or Downlink Control Information (DCI)). Models may not be recognized at the network (NW), and the UE can perform model-level LCM. A UE can have one AI / ML model per function or multiple AI / ML models per function. A function refers to a feature / group of features implemented by AI / ML enabled by configuration, where the configuration is supported based on conditions indicated by the UE's capabilities. In addition to function identification, the UE can also report updates regarding applicable functions within a function.

[0011] In model-based LCM, the model is identified at the NW, and the NW / UE can activate / deactivate / select / switch individual AI / ML models via model ID. The model can be associated with specific configurations / conditions, and additional conditions (e.g., scene, site, dataset) can be determined / identified between the UE and the NW.

[0012] AI / ML model validation is a sub-process of training, such as Figure 2 As shown, the quality of an AI / ML model is evaluated using a dataset different from the dataset used for model training, which helps to select model parameters that generalize from the dataset used for model training.

[0013] AI / ML models can be used for UE localization. Depending on the capabilities, the UE or gNB may have a trained model stored internally, or an untrained AI / ML that can be trained in real time to generate the measurements needed to locate the UE within the radio access network (RAN) coverage area, or to directly predict / determine the UE location by utilizing measurements performed by the UE or gNB on reference signals such as Position Reference Signals (PRS), Sounding Reference Signals (SRS), etc., within the RAN coverage area.

[0014] Measurements predicted / determined by UE using AI / ML models can be defined as, but are not limited to: RSTD It is the reference signal time difference between positioning node j and reference positioning node i. It is measured on the DL PRS signal and always involves two cells (the cells are interchangeably referred to as TRP). UE Rx-Tx time difference It is defined as T UE-Rx -T UE-Tx ; in: ○ TUE-RX It is the timing at which the UE receives downlink subframe #i from the positioning node, defined by the first detected path in time. It is measured on the PRS signal received from the gNB; ○ T UE-TX It is the timing of the uplink subframe #j that the UE sends in time is closest to the subframe #i received from the positioning node.

[0015] gNB measurements, predicted / determined using AI / ML models, can be defined as, but are not limited to: gNB Rx-Tx Time Difference It is defined as T gNB Rx -T gNB-Tx ; in: ○ T gNB-RX The timing of the location node receiving the uplink subframe #i containing the SRS associated with the UE is defined by the first detected path in time. It is measured on the SRS signal received from the UE; ○ T gNB-TX It is the timing of the downlink subframe #j that the positioning node sends in time is closest to the subframe #i received from the UE.

[0016] Timed advance (T) ADV ) It is defined as the time difference T ADV = (T gNB-RX -T gNB-TX ), in: ○ T gNB-RX The timing of the Transmitter-Receiver Point (TRP) receiving the uplink subframe #i containing the PRACH transmitted from the UE is defined by the first path detected in time; ○ T gNB-TX It is the timing of the TRP sending downlink subframe #j that is closest in time to the subframe #i received from the UE; ○ The detected PRACH is used to determine the start of a subframe containing that PRACH; UL Relative Time of Arrival (UL RTOA) It is defined as the start of subframe i of the SRS received in location node j, relative to a configurable reference time. For example, node 1 (e.g., a base station, etc.) measures the reception time of the signal transmitted by the UE relative to the reference time.

[0017] In addition, the UE or gNB can perform power measurements such as Reference Signal Received Power (RSRP) and / or Reference Signal Received Path Power (RSRPP). These measurements can be performed on reference signals such as PRS and SRS.

[0018] Depending on its capabilities, the UE can also perform location measurements on secondary link (SL) resources by utilizing AI / ML models, for example, on SL-PRS transmitted between the target UE and one or more auxiliary or anchor UEs. The UE performing the location measurement is called the target UE, and the UE assisting the target UE in performing the SL location measurement is called the anchor or auxiliary UE.

[0019] In assisted positioning mode, positioning measurements are performed by the UE / gNB and used as input for the AI / ML model residing at the UE / gNB. After completion, the output (including the measurement results) is then reported to the location server. Upon receiving the AI / ML model output (including the measurement results), the location server determines the UE's location within the RAN coverage area. If necessary, the location server can forward the UE's location to another node within the network to facilitate providing the UE's location information to application layers or third parties interested in or requesting to locate the UE within the RAN coverage area for further action. AI / ML model output: New measurements and / or enhancements to existing measurements For example, LOS / NLOS identification, measurement timing and / or angle, and measurement possibilities.

[0020] In direct location mode, measurements performed by the UE / gNB serve as input to the location engine or AI / ML model to predict or determine the UE's location. The AI / ML model resides within the UE node, gNB node, or location server. If necessary, the location server can forward the UE's location to another node within the network to facilitate providing the UE's location information to interested application layers or third parties that have requested to locate the UE within the RAN coverage area for further action. AI / ML model output: UE location For example, fingerprint recognition based on channel observation is used as input to AI / ML models.

[0021] In the Position Measurement Procedure (PMP), the procedure involves performing one or more positioning measurements on the DL RS (e.g., PRS) and / or uplink (UL) RS (e.g., SRS) transmitted between the UE and one or more cells. The cell may also be referred to as the Transmit / Receive Point (TRP) or node. Examples of positioning measurements performed on DL and / or UL signals are RSTD, PRS-RSRP, PRS-RSPP, UE Rx-Tx time difference, Round-Trip Time (RTT), Time of Arrival (TOA), Channel Impulse Response (CIR), Timing Advance (TA), Angle of Departure (AoD), Angle of Arrival (AoA), Power Delay Distribution (PDP), Delay Distribution (DP), etc.

[0022] To perform these measurements, the UE can use a trained model acquired before deployment, or it may need to train its model in real time before performing AI / ML-based positioning measurements to be reported to network nodes (such as location servers). In either case, the UE requires auxiliary information / data from the network to determine or identify how and when to train its AI / ML model, and what information (positioning measurements or estimated location) to report to network nodes (such as location servers).

[0023] Prior to version 16, LTE-based positioning was one of the popular RAT-based positioning solutions. Starting with version 16, positioning is also supported in New Radio (NR). Positioning in NR is subject to... Figure 3 The architecture shown supports this. Interaction between eNodeB 304-2 and gNodeB 304-1 (collectively referred to herein as "Base Station Equipment 304") and UE 306 is supported via the RRC protocol, while the location node interfaces with the UE via the LTE Positioning Protocol (LPP). LPP is a common protocol for both NR and LTE. The Location Management Function (LMF) 302 is the location node in NR. LMF 302 also interacts with Base Station Equipment 304 via the NR Positioning Protocol a (NRPPa).

[0024] Figure 3 The positioning architecture in version 19 will also be used to support AI / ML-based positioning. Version 19 aims to introduce AI / ML-based positioning that will not only utilize traditional protocols but will also rely on predefined / existing reference signals used for positioning.

[0025] The following IE TS 37.355 v 18.0.0 provides positioning output:

[0026] Location service quality (QoS) is typically defined by location accuracy and latency.

[0027] Integrity is something that can be provided by navigation / positioning systems. The accuracy of information of trust Integrity is a metric. For example, integrity includes the system's ability to provide timely warnings to user receivers in the event of a failure. Examples of failure can be found in RAT-independent positioning methods such as assisted GNSS: if a satellite malfunctions, the system should detect it and notify the user not to use the satellite.

[0028] Example Use Cases and Scenarios: Any use case related to location in Ultra-Reliable Low-Latency Communication (URLLC) naturally requires high integrity performance. Some example use cases include V2X, autonomous driving, UAVs (unmanned aerial vehicles), e-health, rail and maritime transport, emergency response, and mission-critical applications. In use cases where large errors can lead to serious consequences, such as health-related impacts, incorrect legal decisions, or incorrect billing calculations, integrity reporting becomes crucial.

[0029] Completeness KPIs and events

[0030] Here are some examples of defined integrity KPIs that can help us identify different integrity events.

[0031] Alarm Limit (AL): The maximum permissible positioning error for integrity purposes. If the positioning error exceeds this limit, the integrity result of the calculated location does not meet the integrity requirements.

[0032] Alarm Time Ago (TTA): The maximum allowed elapsed time from the start of a location failure until the device issues an alarm.

[0033] Integrity risk (IR): refers to the maximum probability of providing a signal that exceeds the tolerance within a given time without warning the user.

[0034] Location integrity: A measure of trust in the accuracy of location-related data and the ability to provide relevant alerts.

[0035] Protection Level (PL): The statistical upper limit of the positioning error (PE), which ensures that, over a time period longer than TTA, the probability per unit time that the true error is greater than AL and PL is less than or equal to AL is less than TIR. That is, PL satisfies the following inequality: Prob per unit of time [(( PE>AL ) & ( PL<=AL )) for longer than TTA ] < TIR Note 1: The PL inequality is valid for all values ​​of AL; Note 2: If the requested TIR cannot be satisfied, the TIR may correspond to an achievable TIR.

[0036] When PL bounds the positioning error in the horizontal plane or on the vertical axis, it is referred to as the Horizontal Protection Level (HPL) or Vertical Protection Level (VPL), respectively. The specific equation for PL is not specified as it is implementation-defined. For PL to be considered valid, it must satisfy the aforementioned inequality.

[0037] Figure 4 shows an example with a Stanford diagram, in which all possible integrity operations and events can be interpreted in its different regions.

[0038] Nominal operation is when PE<PL<AL

[0039] System unavailable is when AL<PL

[0040] Misleading operation is when PL<PE

[0041] Hazardous operation is when PL<AL<PE

[0042] An integrity failure is an integrity event whose duration is longer than TTA and for which no alert is issued within TTA.

[0043] Misleading Information (MI) is an integrity event that occurs when the system declares itself available, and the positioning error exceeds the protection level but does not exceed the alert limit.

[0044] Hazardous Misleading Information (HMI) is an integrity event that occurs when the system declares itself available and the positioning error exceeds the alert limit. SUMMARY OF THE INVENTION

[0045] Various embodiments provide a method for determining whether positioning data should be collected and used as training data for training an Artificial Intelligence / Machine Learning (AI / ML) model. A signaling mechanism may be provided, in which a network node specifies which positioning measurement data should be collected for training purposes and which data should be discarded by providing criteria for data collection to a UE or gNB. Similarly, a UE or gNB may request the data collection it needs with certain quality-related criteria. A first network node may determine a tolerance for ground truth labeling, and then determine one or more reliability criteria for positioning data based on the tolerance. The first network node may provide a request for positioning data and the reliability criteria to a second network node, which will collect the positioning data based on the criteria, and then provide the collected positioning data to the first network node, which may then use the positioning data as training data. The first network node and the second network node may be a UE, a base station apparatus, or a Location Management Function (LMF).

[0046] In one embodiment, a method performed by a second network node to facilitate the collection and use of location data as training data for training an AI / ML model may include: receiving one or more reliability criteria for the location data from a first network node; and collecting location data that satisfies the one or more reliability criteria to obtain the collected location data.

[0047] In another embodiment, the one or more reliability criteria are associated with the confidence level of the location data.

[0048] In another embodiment, the first network node is at least one of a base station, a user equipment (UE), or a location management function (LMF), and the second network node is at least one of a base station, a UE, or an LMF, and the first network node and the second network node are of different types.

[0049] In another embodiment, the one or more reliability criteria are determined based on a predefined configuration at one of the first network node or the second network node.

[0050] In another embodiment, the one or more reliability criteria define limits associated with alarm limits, maximum positioning error, uncertainty level, and confidence level of the positioning data.

[0051] In another embodiment, the alarm limit, the maximum positioning error, the uncertainty level, and the confidence level are based on a geographic region description associated with the positioning data.

[0052] In another embodiment, the one or more reliability criteria are based on measurements obtained from uplink measurements from a UE with a known location.

[0053] In another embodiment, the one or more reliability criteria are based on measurements obtained from downlink measurements at a UE with a known location.

[0054] In another embodiment, the UE is a Positioning Reference Unit (PRU).

[0055] In another embodiment, the method may include: obtaining the measurement result from the UE having the known location, wherein the difference between the measurement result from the UE and the uplink measurement result from the second network node is less than a predefined threshold.

[0056] In one embodiment, the method may include providing the collected location data to the first network node.

[0057] In another embodiment, a second network node for facilitating the collection and use of location data as training data for training an AI / ML model includes processing circuitry that causes the second network node to: receive one or more reliability criteria for location data from a first network node; and collect location data that satisfies the one or more reliability criteria, thereby obtaining the collected location data. In another embodiment, the second network node may perform any of the above embodiments.

[0058] In another embodiment, a method is provided, performed by a first network node, for facilitating the collection and use of location data as training data for training an AI / ML model. The method includes: determining a tolerance for ground truth labeling based on one or more Quality of Service (QoS) criteria; determining one or more reliability criteria for the location data based on the tolerance for ground truth labeling; providing the one or more reliability criteria for the location data to a second network node; and receiving the collected location data from the second network node.

[0059] In one embodiment, the first network node is at least one of a base station, a user equipment (UE), or a location management function (LMF), and the second network node is at least one of a base station, a UE, or an LMF, and the first network node and the second network node are of different types.

[0060] In one embodiment, the QoS standard is received from a client application or from another network node.

[0061] In one embodiment, the one or more reliability criteria are determined based on a predefined configuration at one of the first network node or the second network node.

[0062] In one embodiment, the method further includes: receiving capability information from the second network node, wherein the one or more reliability criteria are based at least in part on the capability information.

[0063] In one embodiment, the one or more reliability criteria define limits associated with alarm limits, maximum positioning error, uncertainty level, and confidence level of the positioning data.

[0064] In one embodiment, the alarm limit, the location error, the uncertainty level, and the confidence level are based on a description of the geographic region associated with the location data.

[0065] In one embodiment, the one or more reliability criteria are based on measurements obtained from uplink measurements from a UE with a known location.

[0066] In one embodiment, the one or more reliability criteria are based on measurements obtained from downlink measurements at a UE with a known location.

[0067] In one embodiment, the UE is a PRU.

[0068] In one embodiment, the second network node obtains the measurement result from the UE having the known location, wherein the difference between the measurement result from the UE and the uplink measurement result from the second network node is less than a predefined threshold.

[0069] In one embodiment, the collected location data includes measurement data and label data, wherein the measurement data of the collected location data meets a first minimum quality standard, and wherein, in response to the label data not meeting a second minimum quality standard, the collected location data is labeled as unlabeled training data, and in response to the label data meeting the second minimum quality standard, the collected location data is labeled as labeled training data.

[0070] In one embodiment, the method further includes: checking the quality of the measurement data in response to the occurrence of one or more of the following: transmitter implementation error; receiver implementation error; radio channel quality between the transmitter and receiver not meeting a predefined threshold; or a predefined radio signal configuration.

[0071] In one embodiment, a first network node is provided for facilitating the collection and use of location data as training data for training an AI / ML model. The first network node includes processing circuitry to cause the network node to: determine a tolerance for ground truth labeling based on one or more Quality of Service (QoS) criteria; determine one or more reliability criteria for the location data based on the tolerance for ground truth labeling; provide the one or more reliability criteria for the location data to a second network node; and receive the collected location data from the second network node. In another embodiment, the first network node may perform any of the above embodiments. Attached Figure Description

[0072] The accompanying drawings, which are incorporated in and form part of this specification, illustrate several aspects of this disclosure and, together with the specification, serve to explain the principles of this disclosure.

[0073] Figure 1 Examples illustrating different stages of an artificial intelligence / machine learning (AI / ML) model training pipeline according to some embodiments of this disclosure are shown. Figure 2Examples of AI / ML model verification according to some embodiments of this disclosure are shown; Figure 3 Examples of positioning architectures according to some embodiments of this disclosure are shown; Figure 4 Examples of Stanford diagrams showing protection levels plotted based on positioning errors are shown according to some embodiments of the present disclosure; Figure 5 A message sequence diagram illustrating a method according to some embodiments of the present disclosure for determining whether location data should be collected and used as training data for training an artificial intelligence / machine learning (AI / ML) model; Figure 6 Examples of communication systems according to some embodiments of this disclosure are shown; Figure 7 User equipment (UE) according to some embodiments of this disclosure is shown; Figure 8 Network nodes according to some embodiments of this disclosure are shown; and Figure 9 This is a block diagram illustrating a virtualized environment in which functionality implemented by some embodiments of the present disclosure can be virtualized according to some embodiments. Detailed Implementation

[0074] The embodiments described below represent information enabling those skilled in the art to practice these embodiments and illustrate the best mode for practicing these embodiments. Upon reading the following description with reference to the accompanying drawings, those skilled in the art will understand the concepts of this disclosure and recognize the application of these concepts not specifically mentioned herein. It should be understood that these concepts and applications fall within the scope of this disclosure.

[0075] Some embodiments contemplated herein will now be described more fully with reference to the accompanying drawings. These embodiments are provided by way of example to convey the scope of the subject matter to those skilled in the art.

[0076] Several challenges exist. One of the main challenges in positioning is identifying the true value. When calculating positioning, there is always associated uncertainty because positioning can be based on triangulation (measurements from at least 3 different Transmitter-Receiver Points (TRPs), satellites, etc.), and the intersection area obtained through time estimation can be regarded as the location of the User Equipment (UE); that is, the location is associated with an area defined by uncertainty and confidence factors.

[0077] The version 19 Work Item Description (WID) for Artificial Intelligence / Machine Learning (AI / ML) has the following components for location: Enhanced positioning accuracy, including [Radio Access Network 1 (RAN1) / RAN2 / RAN3]: a. Direct AI / ML localization: i. (First Priority) Example 1: Based on User Equipment (UE) positioning, use the UE-side model for direct AI / ML positioning. ii. (Second Priority) Example 2b: Positioning based on UE Assisted / Location Management Function (LMF), using the LMF-side model for direct AI / ML positioning. iii. (First Priority) Example 3b: NG-RAN Node-Assisted Localization, using LMF-side model for direct AI / ML localization b. AI / ML-assisted localization i. (Second Priority) Example 2a: UE-assisted / LMF-based localization, utilizing UE-side models and AI / ML-assisted localization. ii. (First priority) Example 3a: NG-RAN node-assisted localization, using gNB side model, AI / ML-assisted localization.

[0078] As suggested by WID, different combinations are possible; that is, there could be five different instances. The model could reside in a single node (Location Management Function (LMF), UE, gNB), and this node would need to rely on another node in the network for data collection to train its AI / ML model. For example, the AI / ML model might reside in the LMF, and the gNB and UE must collect data and provide it to the LMF for AI / ML model training; the data reported by the UE / gNB may contain some degree of error. Training AI / ML at the LMF with such data would lead to incorrect inferences when predicting UE location.

[0079] For AI / ML-based localization, regardless of the node where the AI / ML model is deployed, performing labeled data collection with higher accuracy becomes particularly important. The labels of the data collected for training the AI / ML model should maintain a certain level of accuracy. More importantly, if the data is collected for AI / ML model training, the labels or ground truth values ​​corresponding to the collected data should also reach a certain level of accuracy.

[0080] The UE (especially) is unclear about which data can be safely collected for training purposes and which data should be discarded, or basically what is acceptable tag data for the UE to apply to the associated / acquired measurements.

[0081] For AI / ML-assisted localization use cases, when the model output is an intermediate localization measurement, the ground truth label is the actual measurement output, such as Time of Arrival (TOA), Angle of Arrival (AOA), Angle of Departure (AOD), Reference Signal Time Difference (RSTD), Reference Signal Received Power (RSRP), Reference Signal Received Path Power (RSRPP), gNB Receive-Transmit (Rx-Tx) Time Difference, UE Rx-Tx Time Difference, Timing Advance, or True Loss Classification. When the model output is a location, the ground truth label is the actual UE location. It is unclear what input data can be used, and how this data should be collected to obtain labels for the actual location (direct localization) or for the actual localization measurement (assisted localization).

[0082] Certain aspects of this disclosure and its embodiments may provide solutions to these or other challenges.

[0083] Various embodiments provide a method for determining whether location data should be collected and used as training data for training an AI / ML model. A signaling mechanism can be provided in which a network node specifies which location measurement data should be collected for training purposes and which should be discarded by providing criteria for data collection to a UE or gNB. Similarly, a UE or gNB can request data collection that meets certain quality-related criteria. A first network node can determine a tolerance for ground truth labeling and then, based on that tolerance, determine one or more reliability criteria for the location data. The first network node can provide a request for location data and reliability criteria to a second network node, which will collect the location data based on the criteria and then provide the collected location data to the first network node, which can then use the location data as training data. The first and second network nodes can be a UE, base station equipment, or a location management function (LMF).

[0084] This disclosure provides a signaling mechanism in which an NW node specifies which positioning measurement data (e.g., CIR, Power Delay Distribution (PDP), RSRP, RSRPP, Time of Arrival, RSTD, Location) should be collected for training purposes and which data should be discarded by providing the UE or gNB with criteria for data collection. Similarly, the UE or gNB can request data collection that it requires based on specific quality-related criteria.

[0085] A basic invention flow is proposed, wherein node 1 (e.g., a first network node) and node 2 (a second network node) can be a UE, gNB (base station equipment), or LMF.

[0086] Node 1 can determine the tolerance for the true value based on its own knowledge (e.g., predefined information, preconfigured information) or based on the configuration information, the location service quality or integrity service requirements it receives from the LCS client (or other nodes).

[0087] Node 1 prepares the quality standards for the true values.

[0088] Node 2 follows logic to collect or discard measurements based on the obtained positioning error or measurement error and compare it with a threshold.

[0089] The collected results are stored by the training data collection entity. The training data collection entity can have many variations, such as node 1 or node 2, or a node that hosts the model used for model inference, or a model training entity for a given AI / ML model.

[0090] Certain embodiments may provide one or more of the following technical advantages. Location requests originate from different clients, each with its own requirements. Some may have very strict requirements for location accuracy, while others may have a more lenient tolerance for the resulting location accuracy. For AI / ML-based location, different datasets can be collected based on the classification of location requirements. One dataset can be used for high-accuracy location, while another dataset can be used to improve location accuracy in very chaotic environments to achieve perfect accuracy. The mechanisms in this disclosure allow entities to train their AI / ML models according to their preferences (Quality of Service (QoS) requirements for location) and set criteria for collecting ground truth values ​​from data.

[0091] The terminology label data used in this article is the true value for positioning in terms of real location coordinates or real location measurements, without any error.

[0092] When the network (NW) provides auxiliary data for integrity, it can limit the error. In this case, the NW specifies alarm limits to configure true values. This means that if the maximum positioning error (PE) is within the specified alarm limits, the UE should consume the information (use) for data collection; otherwise, the UE should discard the data (e.g., measurements, location).

[0093] When the AI / ML model is in the LMF and the LMF needs to configure the UE for data collection, the following can be part of the LPP auxiliary data; or when the TRP / gNB must collect data for the AI / ML model deployed in the LMF, the following can be in the NR Positioning Protocol a (NRPPa); or when the AI / ML model is in the LMF and the LMF needs to configure the TRP / gNB for data collection, the following can be in the NRPPa.

[0094] Similarly, when the AI / ML model is in the UE and the UE needs tag data collected by the LMF to train its AI / ML model, the following can be part of an LPP supplementary data request from the UE to the LMF. The UE can then specify the criteria for tag data collection. This applies to scenarios where the UE has an LCS client and the QoS is known to the UE. Data collection requirements for the LMF can also be set via a supplementary request (Location Services LCS message), using LPP for the LMF, or via Radio Resource Control (RRC) for the gNB.

[0095] The same applies when the gNB has an AI / ML model and needs to collect tag data from the UE. It can be configured via RRC for the UE. The same applies when the gNB has an AI / ML model and needs to collect tag data (real values) from the LMF, and in this case, it can be configured via NRPPa for the LMF.

[0096] Below is an example of a signaling method used to provide a location estimation threshold. It will be understood that other variations can also be used, for example, other maximum acceptable errors besides 20 meters:

[0097] This field provides a threshold for acceptable location estimation error, where the location information was used in the training data collection for the AI / ML model used for localization: horizontalAlertLimit A threshold for the horizontal position estimation error along the semi-major axis of the error ellipse is provided. Scale factor: 0.01 meters; range: 0-20 meters. verticalAlertLimit A threshold for vertical position estimation error is provided. Scale factor: 0.01 meters; range: 0-20 meters.

[0098] Another approach, or a method combined with the above, specifies whether the uncertainty is within a certain limit and whether the associated measurement confidence factor is above a certain threshold.

[0099] The UE can support different geographic shapes for location reporting formats, so the requesting entity can receive this capability and provide uncertainty and confidence factors corresponding to the supported location format (Geographic Area Description (GAD) shape).

[0100] Here is an example:

[0101] In IE, the field UncertaintyThreshold for DataCollection corresponds to one of the location reporting formats (geographic shapes) supported by the node (UE, gNB, LMF) and the uncertainty threshold used for the corresponding location reporting format. Similarly, IE ConfidenceThresholdsForDataCollection corresponds to the confidence levels specified for horizontal and vertical directions, depending on the node's capabilities.

[0102] Two possible options for setting criteria to check whether a measurement can be used as a true value: Criteria for checking the accuracy of intermediate measurements: In this case, thresholds are set for measurement errors (such as TOA, AOA, AOD, LOS / NLOS probability, CIR, PDP, RSRP, RSRPP, RSRD, UE Rx-Tx time difference, gNBRx-Tx time difference, TA). Criteria for checking positioning accuracy: In this case, a threshold is set for the positioning error.

[0103] In some embodiments, criteria for the true value measurement are obtained based on measurements from UL measurements from UEs (e.g., PRUs or others) with known locations (stored at network node 1 in the main graph). Since the location of the UE can be defined by an uncertainty / confidence parameter, the true value UL measurement can be associated with an uncertainty / confidence interval.

[0104] In some other embodiments, criteria for measuring the true value are obtained based on measurements from DL measurements at a UE (e.g., PRU) with a known location.

[0105] (According to TS 38.305 (Section 5.4.5), the location server can compare the PRU measurement with the measurement expected at the known PRU location to determine correction terms for other nearby target devices. The DL and / or UL location measurements for other target devices can then be corrected based on the previously determined correction terms. Therefore, if the UE is not a PRU, the DL and / or UL location measurements for other target devices can be corrected based on the previously determined correction terms.)

[0106] In some other embodiments, network node 2 is a UE located near a UE with a known location within a predefined confidence interval. If the difference between these measurements and measurements obtained from a UE with a known location is less than a predefined threshold of network node 1, the UL measurement can be considered a true value.

[0107] Generally, when using supervised learning in model training, a training data sample consists of two parts [X] sample (j), Y sample (j)] Composition: (a) Measurement data X corresponding to the model input X sample (j); (b) Label data Y corresponding to the model output Y sample (j). When label data is unavailable or unacceptable, the training data samples consist only of the measurement data X corresponding to the model input X. sample (j) is composed of, and this is an unlabeled training data sample [X] sample (j)]. Unlabeled training data samples can be used for semi-supervised or unsupervised learning during the training process. Note that here, X sample (j) and Y sample (j) are typically vectors or matrices (i.e., not limited to single values), whose elements can be real or complex numbers, depending on the model input and output design choices.

[0108] When collecting training data samples, it is necessary to examine the measurement data X. sample (j) and label data Y sample (j) Both, and only data of the required quality should be stored as part of the training dataset.

[0109] If the sample measurement data X sample (j) If the minimum quality standard is not met, then even if the corresponding label data Y sample (j) Even if the sample has sufficient quality, it should still be rejected.

[0110] If the sample measurement data X sample (j) If a sample meets the minimum quality criteria for X, then it can be accepted as a training data sample. If the corresponding label data Y... sample (j) If the minimum quality criteria for Y are available and met, then labeled training data samples can be stored: [X] sample (j), Y sample (j)]; otherwise, unlabeled training data samples can be stored: [X sample (j)].

[0111] Used to check measurement data X sample (j) The quality criteria may include one or more of the following aspects: Transmitter implementation error: c. For example, if the synchronization error between TRPs is too large, the measurement data of the downlink signal may not be of sufficiently good quality; d. In another example, if the timing error transmitted by the UE is too large, the measurement data of the uplink signal may not be of sufficiently good quality; Receiver implementation error: e. For example, if the channel estimation error at the receiver is too large, the measurement data may not be of sufficiently good quality; Radio channel quality between transmitter and receiver: f. For example, the SINR of the received signal used to generate measurement data may be too low to be acceptable; Radio signal configuration, in which radio signals are used to acquire measurement data: g. For example, if the reference signal bandwidth is configured too small, the measurement data obtained from the reference signal will not have sufficient quality. Typical reference signals used for positioning include DL-PRS for downlink and SRS for uplink.

[0112] In another embodiment, Node 2 may report its ability to collect data regarding the accuracy of potential true value measurements to Node 1, and this report may be based on a capability request from Node 1. In this case, Node 2 may or may not be a PRU. For example, if Node 2 is not a PRU, but its accuracy in true value measurements meets (or substantially meets) Node 1's requirements, then Node 1 may configure Node 2 to report the corresponding true value measurements.

[0113] NW nodes (e.g., LMFs) can also impose requirements on a set of capabilities to ensure that only those UEs that can meet or possess the capabilities can thus be considered as data sources for AIML purposes (i.e., capable of providing location measurements and / or estimated locations or capable of transmitting SRS). For example, if the UE supports: Timing error group characteristics (as defined in TS 38.305v 18.0.0): One definition is as follows: h. Tx Timing Error: The result of Tx time delay involved in signal transmission. It is the uncalibrated Tx time delay involved in the transmission of DL-PRS / UL SRS signals, or the residual delay after Tx time delay internal calibration / compensation within the TRP / UE. This calibration / compensation may also include calibration / compensation of relative time delays between different RF chains within the same TRP / UE, and may also take into account the offset of the Tx antenna phase center from the physical antenna center; It features a GNSS chipset (i.e., it can use high-accuracy positioning HA-GNSS (TS 37.355 v18.0.0) as a location source). The UE knows its known location (e.g., if the UE is stationary / fixed) or moves / jumps within the known location; If the UE can operate in a UE-based positioning mode (as defined in TS 37.355v 18.0.0).

[0114] In one approach, before configuring SRS resources for data collection at the gNB side for the UE, the gNB can determine whether the UE can provide true values ​​related to the estimated location with a certain level of accuracy. This information can be provided by the UE in UE capability signaling, or in response to a gNB or LMF request. If the accuracy indicated by the UE is higher than a certain threshold, the gNB can configure the SRS resources for the UE for data collection purposes. The UE can determine whether it can meet the accuracy requirement based on previous measurements performed from the same location (beam, cell) and indicate to the NW whether it can meet the requirement. The UE then sends the UL along with the true value estimate (e.g., location coordinates in the shape of the geographic area description GAD) to the gNB. The gNB can also indicate the expected level of accuracy so that, in this case, the UE continuously detects that the true value estimate no longer falls within the configured accuracy indicated by the UE to the gNB. The gNB can then deconfigure the UE with the UL SRS transmission, or it can instruct the UE to temporarily disable the UL SRS transmission.

[0115] In another example, the gNB can determine whether the UE can provide a true value related to the estimated location with a specific accuracy from the LMF. For example, the gNB can request the LMF whether the UE can provide a true value with a specific accuracy. Alternatively, the LMF can inform the gNB whether the UE can provide a true value with a specific accuracy. In this case, the gNB can configure the UL SRS resources required for data collection on the NW side for the UE.

[0116] Similarly, if the LMF must collect data from the UE for AIML data collection for training purposes, it can only request to send a PRS to the gNB if it knows that the UE can act as a data source, i.e., the UE is able to generate measurements and location estimates that satisfy the true value criterion.

[0117] When a criterion for the truth value has been provided from one node to another, and the receiving node cannot meet the requester's truth value criterion, it can provide an error or failure message with a reason code, as shown in the following example: "LocationEstimateWithGivenAccuracyNotPossible", "failedToEstimateLocation", "groundTruthNotDetermined".

[0118] In addition, nodes can include certain location measurements and implemented location satisfaction criteria, and can be stored for future use. This can also be done if normal positioning toward the target UE is already configured, and if the target UE is a cooperative UE (willing to provide both location and measurements) and satisfies the true value criterion.

[0119] Figure 5 A message sequence diagram illustrating a method for determining whether location data should be collected and used as training data for training an artificial intelligence / machine learning (AI / ML) model, according to some embodiments of the present disclosure.

[0120] Figure 5 The method can begin at step 508, where a first network node (e.g., node 1) sends a capability request message to a second network node 504 (e.g., node 2). In response, at step 510, the second network node 504 can respond to the first network node 502 with a capability report.

[0121] In step 512, the first network node 502 may determine the tolerance for ground truth labeling based on one or more QoS standards. In one embodiment, the QoS standards are received from a client application or from another network node.

[0122] In step 514, the first network node 502 determines one or more reliability criteria for the location data based on tolerances used for ground truth labeling. In one embodiment, the one or more reliability criteria define constraints associated with the uncertainty level and confidence level of the location data. In another embodiment, one or more reliability criteria are determined based on a predefined configuration at either the first network node 502 or the second network node 504. In one embodiment, the uncertainty level and confidence level are based on a geographic region description associated with the location data. In another embodiment, one or more reliability criteria are based on measurements obtained from uplink measurements from a UE 506 with a known location. In one embodiment, one or more reliability criteria are based on measurements obtained from downlink measurements at a UE 506 with a known location. In one embodiment, the UE may be a PRU.

[0123] In one embodiment, the second network node obtains the measurement result from the UE 506 with a known location in step 520, wherein the difference between the measurement result from the UE and the uplink measurement result from the second network node (504) is less than a predefined threshold.

[0124] In step 516, the first network node 502 provides the second network node 504 with one or more reliability criteria for locating data.

[0125] In step 518, the second network node 504 collects location data that meets one or more reliability criteria, thereby obtaining the collected location data. Location data that does not meet one or more reliability criteria may be discarded. The second network node 504 may also optionally obtain measurement data from the UE 506 that can be collected as location data.

[0126] In step 522, the second network node 504 can provide location data to the first network node 502.

[0127] In step 524, the first network node 502 may optionally check the quality of the measurement data in response to one or more of the following: transmitter implementation error; receiver implementation error; radio channel quality between the transmitter and receiver not meeting a predefined threshold; or predefined radio signal configuration.

[0128] In step 526, the first network node 502 can use localization data, including both measurement data and label data, as unlabeled training data or labeled training data to train an AI / ML model at the first network node 504.

[0129] Figure 6 An example of a communication system 600 according to some embodiments is shown.

[0130] In this example, communication system 600 includes telecommunications network 602, which includes access network 604 such as a radio access network (RAN) and core network 606. Core network 606 includes one or more core network nodes 608. Access network 604 includes one or more access network nodes, such as network nodes 610A and 610B (one or more of which may generally be referred to as network node 610), or any other similar 3GPP access node or non-3GPP access point (AP). Furthermore, those skilled in the art will understand that network nodes are not necessarily limited to implementations in which the radio and baseband portions are provided and integrated by a single vendor. Therefore, it will be understood that network nodes include decomposed implementations or portions thereof. For example, in some embodiments, telecommunications network 602 includes one or more Open RAN (ORAN) network nodes. An ORAN network node is a node in telecommunications network 602 that supports ORAN specifications (e.g., specifications published by the O-RAN Alliance or any similar organization) and can operate independently or in conjunction with other nodes to implement one or more functions of any node in telecommunications network 602 (including one or more network nodes 610 and / or core network node 608). Network node 610 may be an example of a gNB implementation of the first network node 502 or the second network node 504. Similarly, when core network node 608 is an implementation of the first network node 502 or the second network node 504, it can implement location management functions.

[0131] Examples of ORAN network nodes include Open Radio Units (O-RUs), Open Distributed Units (O-DUs), Open Central Units (O-CUs) (including O-CU control planes (O-CU-CPs) or O-CU user planes (O-CU-UPs)), RAN Intelligent Controllers (near real-time or non-real-time) with managed software or software plugins (such as near real-time control applications (e.g., xApps) or non-real-time control applications (e.g., rApps)), or any combination thereof (the adjective "open" specifies support for the ORAN specification). Network nodes can support the specification by, for example, supporting interfaces defined by the ORAN specification, such as A1, F1, W1, E1, E2, X2, Xn interfaces, Open Fronthaul User Plane interfaces, or Open Fronthaul Management Plane interfaces. Furthermore, ORAN access nodes can be logical nodes within physical nodes. Additionally, ORAN network nodes can be implemented in a virtualized environment (described further below) where one or more network functions are virtualized. For example, the virtualized environment can include an O-Cloud computing platform orchestrated by a service management and orchestration framework via the O-2 interface or similar technologies defined by the O-RAN Consortium. Network node 610 facilitates direct or indirect connections of user equipment (UE), such as connecting UE 612A, 612B, 612C, and 612D (one or more of which may generally be referred to as UE 612) to core network 606 via one or more wireless connections. UE 612 may be an implementation of UE when the first network node 502 or the second network node 504 is a UE.

[0132] Examples of wireless communication via wireless connection include sending and / or receiving wireless signals using electromagnetic waves, radio waves, infrared waves, and / or other types of signals suitable for transmitting information without the use of wires, cables, or other conductors. Furthermore, in various embodiments, communication system 600 may include any number of wired or wireless networks, network nodes, UEs, and / or any other components or systems that can facilitate or participate in the communication of data and / or signals (whether via wired or wireless connections). Communication system 600 may include any type of communication, telecommunications, data, cellular, radio network, and / or other similar type of system and / or interface with it.

[0133] UE 612 can be any of a variety of communication devices, including wireless devices that are set up, configured, and / or operable to communicate wirelessly with network node 610 and other communication devices. Similarly, network node 610 is set up, capable of, configured, and / or operable to communicate directly or indirectly with UE 612 and / or other network nodes or devices in telecommunication network 602 to enable and / or provide network access (such as wireless network access) and / or perform other functions (such as management) in telecommunication network 602.

[0134] In the example shown, core network 606 connects network node 610 to one or more hosts, such as host 616. These connections can be direct or indirect, via one or more intermediate networks or devices. In other examples, network nodes can be directly coupled to hosts. Core network 606 includes one or more core network nodes (e.g., core network node 608) composed of hardware and software components. The characteristics of these components can be substantially similar to those described for UEs, network nodes, and / or hosts, such that the description generally applies to the corresponding components of core network node 608. Example core network nodes include one or more of the following functions: Mobile Switching Center (MSC), Mobility Management Entity (MME), Home Subscriber Server (HSS), Access and Mobility Management Function (AMF), Session Management Function (SMF), Authentication Server Function (AUSF), Subscription Identifier Dehiding Function (SIDF), Unified Data Management (UDM), Security Edge Protection Agent (SEPP), Network Open Function (NEF), and / or User Plane Function (UPF).

[0135] Host 616 may be under the ownership or control of a service provider other than the operator or provider of access network 604 and / or telecommunications network 602, and may be operated by or on behalf of the service provider. Host 616 may host various applications to provide one or more services. Examples of such applications include real-time and pre-recorded audio / video content, data collection services (such as acquiring and editing data on various environmental conditions detected by multiple UEs), analytics functions, social media, functions for controlling or otherwise interacting with remote devices, functions for alarms and monitoring centers, or any other such functions performed by the server.

[0136] Overall, Figure 6The communication system 600 enables connections between the UE, network nodes, and the host. In this sense, the communication system can be configured to operate according to predefined rules or procedures (such as specific standards), including but not limited to: Global System for Mobile Communications (GSM); Universal Mobile Telecommunications System (UMTS); Long Term Evolution (LTE), and / or other suitable second, third, fourth, or fifth generation (2G, 3G, 4G, or 5G) standards, or any applicable future generation standard (e.g., sixth generation (6G)); Wireless Local Area Network (WLAN) standards, such as the Institute of Electrical and Electronics Engineers (IEEE) 802.11 standard (WiFi); and / or any other suitable wireless communication standards, such as Global Microwave Access Interoperability (WiMax), Bluetooth, Z-wave, Near Field Communication (NFC), ZigBee, LiFi, and / or any Low Power Wide Area Network (LPWAN) standards, such as LoRa and Sigfox.

[0137] In some examples, telecommunications network 602 is a cellular network implementing 3GPP standardized features. Therefore, telecommunications network 602 can support network slicing to provide different logical networks to different devices connected to it. For example, telecommunications network 602 can provide ultra-reliable low-latency communication (URLLC) services to some UEs, enhanced mobile broadband (eMBB) services to others, and / or massive machine-type communication (mMTC) / massive IoT services to still others.

[0138] In some examples, UE 612 is configured to send and / or receive information without direct human interaction. For example, the UE may be designed to send information to access network 604 according to a predetermined schedule when triggered by an internal or external event or in response to a request from access network 604. Additionally, the UE may be configured to operate in a single radio access technology (RAT) or multiple radio access technology (RAT) or multiple standards mode. For example, the UE may operate using any one or a combination of Wi-Fi, New Radio (NR), and LTE, i.e., configured for multiple radio dual connectivity (MR-DC), such as Evolved UMTS Terrestrial RAN (E-UTRAN) NR-Dual Connectivity (EN-DC).

[0139] In this example, hub 614 communicates with access network 604 to facilitate indirect communication between one or more UEs (e.g., UE 612C and / or 612D) and network nodes (e.g., network node 610B). In some examples, hub 614 may be a controller, router, content source and analytics node, or any other communication device described herein relating to the UE. For example, hub 614 may be a broadband router that enables the UE to access core network 606. As another example, hub 614 may be a controller that sends commands or instructions to one or more actuators in the UE. Commands or instructions may be received from the UE, network node 610, or via executable code, scripts, procedures, or other instructions in hub 614. As another example, hub 614 may be a data collector that acts as a temporary storage device for UE data, and in some embodiments, data analysis or other processing may be performed. As another example, hub 614 may be a content source. For example, for a UE acting as a virtual reality (VR) headset, display, speaker, or other media delivery device, hub 614 can acquire VR assets, video, audio, or other media or data related to sensed information via a network node, and then provide them to the UE directly, after performing local processing, and / or after adding additional local content. In yet another example, hub 614 acts as a proxy server or orchestrator for the UE, particularly when one or more UEs are low-power IoT devices.

[0140] Hub 614 may have a constant / persistent or intermittent connection to network node 610B. Hub 614 may also allow different communication schemes and / or scheduling between hub 614 and UEs (e.g., UEs 612C and / or 612D) and between hub 614 and core network 606. In other examples, hub 614 is connected to core network 606 and / or one or more UEs via a wired connection. Furthermore, hub 614 may be configured to connect to a machine-to-machine (M2M) service provider via access network 604 and / or to another UE via a direct connection. In some scenarios, a UE may establish a wireless connection with network node 610 while still being connected via hub 614 through a wired or wireless connection. In some embodiments, hub 614 may be a dedicated hub, that is, a hub whose primary function is to route communication from network node 610B to UE / from UE to network node 610B. In other embodiments, hub 614 may be a non-dedicated hub, that is, a device capable of operating to route communication between the UE and network node 610B, but also capable of operating as a communication start and / or end point for a specific data channel.

[0141] Figure 7A UE 700 according to some embodiments is illustrated. As used herein, a UE refers to a device capable of, configured to, set up, and / or operable to wirelessly communicate with network nodes and / or other UEs. Examples of UEs include, but are not limited to, smartphones, mobile phones, cellular phones, IP-based voice (VoIP) phones, wireless local loop phones, desktop computers, personal digital assistants (PDAs), wireless cameras, game consoles or devices, music storage devices, playback devices, wearable terminal devices, wireless endpoints, mobile stations, tablet computers, laptops, laptop embedded devices (LEEs), laptop mounted devices (LMEs), smart devices, wireless client devices (CPEs), vehicles, in-vehicle or in-vehicle embedded / integrated wireless devices, etc. Other examples include any UE identified by 3GPP, including narrowband Internet of Things (NB-IoT) UEs, machine-type communication (MTC) UEs, and / or enhanced MTC (eMTC) UEs.

[0142] The UE may support device-to-device (D2D) communication, for example, by implementing 3GPP standards for secondary link communication, dedicated short-range communication (DSRC), vehicle-to-vehicle (V2V), vehicle-to-infrastructure (V2I), or vehicle-to-everything (V2X). In other examples, the UE may not necessarily have a user in the sense of a human user who owns and / or operates the associated equipment. Instead, the UE may represent a device intended for sale to or operated by a human user but which may not, or initially may not, be associated with a particular human user (e.g., a smart sprinkler controller). Alternatively, the UE may represent a device not intended for sale to or operated by an end user but which may be associated with or operated for the benefit of a user (e.g., a smart power meter).

[0143] UE 700 includes processing circuitry 702, which is operatively coupled via bus 704 to input / output interface 706, power supply 708, memory 710, communication interface 712, and / or any other component or any combination thereof. A particular UE may utilize... Figure 7 All components or subsets of components are shown. The level of integration between components can vary depending on the UE. Furthermore, a particular UE can contain multiple instances of components, such as multiple processors, memories, transceivers, transmitters, receivers, etc.

[0144] Processing circuitry 702 is configured to process instructions and data and can be configured to implement any sequential state machine operable to execute instructions of a machine-readable computer program stored in memory 710. Processing circuitry 702 can be implemented as one or more hardware-implemented state machines (e.g., using discrete logic, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), etc.); programmable logic and suitable firmware; one or more stored computer programs, general-purpose processors (such as microprocessors or digital signal processors (DSPs)) and suitable software; or any combination thereof. For example, processing circuitry 702 may include multiple central processing units (CPUs).

[0145] In this example, the input / output interface 706 can be configured to provide one or more interfaces to input devices, output devices, or one or more input and / or output devices. Examples of output devices include speakers, sound cards, video cards, displays, monitors, printers, actuators, transmitters, smart cards, another output device, or any combination thereof. Input devices can allow users to capture information into the UE 700. Examples of input devices include touch-sensitive or presence-sensitive displays, cameras (e.g., digital cameras, digital video cameras, webcams, etc.), microphones, sensors, mice, trackballs, steering wheels, scroll wheels, smart cards, etc. Presence-sensitive displays may include capacitive or resistive touch sensors to sense input from the user. Sensors may be, for example, accelerometers, gyroscopes, tilt sensors, force sensors, magnetometers, optical sensors, proximity sensors, biometric sensors, etc., or any combination thereof. Output devices can use the same type of interface port as input devices. For example, a Universal Serial Bus (USB) port can be used to provide both input and output devices.

[0146] In some embodiments, power supply 708 is configured as a battery or battery pack. Other types of power sources may be used, such as external power sources (e.g., power outlets), photovoltaic devices, or power units. Power supply 708 may also include power circuitry for delivering power from power supply 708 itself and / or external power sources to various parts of UE 700 via input circuitry or interfaces (such as power cables). The delivery of power may, for example, be used to charge power supply 708. The power circuitry may perform any formatting, conversion, or other modifications to the power from power supply 708 to suit the appropriate components of UE 700 to which power is supplied.

[0147] Memory 710 may be, or may be configured to include, memory such as random access memory (RAM), read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), disk, optical disk, hard disk, removable cassette tape, flash drive, etc. In one example, memory 710 includes one or more applications 714 (such as operating system, web browser application, widget, utility engine, or other application) and corresponding data 716. Memory 710 can store any one or a combination of various operating systems for use by UE 700.

[0148] The memory 710 can be configured to include multiple physical drive units, such as a redundant array of independent disks (RAID), flash memory, a USB flash drive, an external hard drive, a thumb drive, a pen drive, a key drive, a high-density digital versatile optical disc (HD-DVD) drive, an internal hard drive, a Blu-ray disc drive, a holographic digital data storage (HDDS) disc drive, an external mini dual in-line memory module (DIMM), synchronous dynamic random access memory (SDRAM), external micro DIMM SDRAM, smart card memory (such as a tamper-proof module in the form of a universal integrated circuit card (UICC), including one or more subscriber identification modules (SIMs), such as USIM and / or ISIM), other memory, or any combination thereof. The UICC can be, for example, an embedded UICC (eUICC), an integrated UICC (iUICC), or a removable UICC, often referred to as a "SIM card." The memory 710 can allow the UE 700 to access instructions, applications, etc., stored on transient or non-transient storage media to offload or upload data. Articles manufactured using communication systems may be tangibly embodied in or contained in memory 710, which may be or include a device-readable storage medium.

[0149] Processing circuitry 702 can be configured to communicate with an access network or other network using communication interface 712. Communication interface 712 may include one or more communication subsystems and may include or be communicatively coupled to antenna 722. Communication interface 712 may include one or more transceivers for communication, such as communication with one or more remote transceivers capable of wireless communication (e.g., another UE or a network node in the access network). Each transceiver may include a transmitter 718 and / or a receiver 720 adapted to provide network communication (e.g., optical, electrical, frequency allocation, etc.). Furthermore, transmitter 718 and receiver 720 may be coupled to one or more antennas (e.g., antenna 722) and may share circuitry, software, or firmware, or alternatively, may be implemented separately.

[0150] In the illustrated embodiment, the communication functions of the communication interface 712 may include cellular communication, Wi-Fi communication, LPWAN communication, data communication, voice communication, multimedia communication, short-range communication such as Bluetooth, NFC, location-based communication such as using a Global Positioning System (GPS) to determine location, another similar communication function, or any combination thereof. Communication may be implemented according to one or more communication protocols and / or standards, such as IEEE 802.11, Code Division Multiple Access (CDMA), Wideband CDMA (WCDMA), GSM, LTE, New Radio (NR), UMTS, WiMax, Ethernet, Transmission Control Protocol / Network Protocol (TCP / IP), Synchronous Optical Network (SONET), Asynchronous Transfer Mode (ATM), Fast User Datagram Protocol Internet Connection (QUIC), Hypertext Transfer Protocol (HTTP), etc.

[0151] Regardless of the sensor type, the UE can provide the output of data captured by its sensors via its communication interface 712 through a wireless connection with a network node. Data captured by the UE's sensors can be transmitted via another UE through the wireless connection with the network node. The output can be periodic (e.g., every 15 minutes if it reports the sensed temperature), random (e.g., to balance the reporting load from multiple sensors), responsive to a triggered event (e.g., sending an alarm when humidity is detected), responsive to a request (e.g., a user-initiated request), or a continuous stream (e.g., real-time video feed of a patient).

[0152] As another example, the UE includes an actuator, motor, or switch associated with a communication interface configured to receive wireless input from a network node via a wireless connection. The state of the actuator, motor, or switch can change in response to the received wireless input. For example, the UE may include a motor that adjusts the control surfaces or rotors of a flying drone based on the received input, or control a robotic arm performing a medical procedure based on the received input.

[0153] When taking the form of an IoT device, the UE can be a device for one or more application areas, including but not limited to urban wearable technology, extended industry applications, and healthcare. Non-limiting examples of such IoT devices include devices that are or are embedded in the following: connected refrigerators or freezers, televisions, connected lighting devices, electricity meters, robotic vacuum cleaners, voice-controlled smart speakers, home security cameras, motion detectors, thermostats, smoke detectors, door / window sensors, flood / humidity sensors, electric door locks, connected doorbells, air conditioning systems such as heat pumps, autonomous vehicles, surveillance systems, weather monitoring devices, vehicle parking monitoring devices, electric vehicle charging stations, smartwatches, fitness trackers, head-mounted displays for augmented reality (AR) or VR, wearable devices for haptic or sensory enhancement, sprinklers, animal or object tracking devices, sensors for monitoring plants or animals, industrial robots, unmanned aerial vehicles (UAVs), and any kind of medical device, such as heart rate monitors or remote-controlled surgical robots. The UE in the form of an IoT device includes the circuitry and / or software associated with the intended application of the IoT device and for... Figure 7 Other components described in UE 700 shown.

[0154] As another specific example, in IoT scenarios, a UE can represent a machine or other device that performs monitoring and / or measurement and sends the results of such monitoring and / or measurement to another UE and / or network node. In this case, the UE can be an M2M device, which can be referred to as an MTC device in the 3GPP context. As a specific example, the UE can implement the 3GPP NB-IoT standard. In other scenarios, a UE can represent a vehicle (such as a car, bus, truck), ship, aircraft, or other equipment capable of monitoring and / or reporting its operational status or other functions associated with its operation.

[0155] In practice, any number of UEs can be used together for a single use case. For example, the first UE may be or be integrated into the drone and provide the drone's speed information (obtained via a speed sensor) to a second UE, which acts as a remote controller for operating the drone. When the user makes changes from the remote controller, the first UE can adjust the throttle on the drone (e.g., by controlling the actuators) to increase or decrease the drone's speed. The first UE and / or the second UE may also include multiple functions described above. For example, the UE may include sensors and actuators and handle data communication between the speed sensor and the actuators.

[0156] Figure 8 A network node 800 according to some embodiments is illustrated. As used herein, a network node refers to a device that is capable of, configured, set up, and / or operable to communicate directly or indirectly with a UE and / or other network nodes or devices in a telecommunications network. Examples of network nodes include, but are not limited to, APs (e.g., radio APs), base stations (BSs) (e.g., radio BSs, Node Bs, evolved Node Bs (eNBs), NR Node Bs (gNBs)), and O-RAN nodes or components of O-RAN nodes (e.g., O-RUs, O-DUs, O-CUs).

[0157] Base stations can be classified based on the coverage they provide (or, in other words, their transmit power level), and thus, depending on the coverage provided, they can be called femtocells, picocells, microcells, or macrocells. A base station can be a relay node or a relay donor node controlling a relay. A network node can also include one or more (or all) portions of a distributed radio base station (such as centralized digital units, distributed units (e.g., in O-RAN access nodes), and / or remote radio units (RRUs) (sometimes referred to as remote radio headends (RRHs)). Such RRUs may or may not be integrated with an antenna as antenna-integrated radios. A portion of a distributed radio base station can also be referred to as a node in a distributed antenna system (DAS).

[0158] Other examples of network nodes include multi-transmission point (multi-TRP) 5G access nodes, multi-standard radio (MSR) equipment (such as MSR BS), network controllers such as radio network controllers (RNC) or BS controllers (BSC), base transceiver stations (BTS), transmission points, transmission nodes, multi-cell / multicast coordination entities (MCE), operation and maintenance (O&M) nodes, operation support system (OSS) nodes, self-organizing network (SON) nodes, location nodes (e.g., evolved serving mobile location center (E-SMLC)), and / or minimized drive test (MDT).

[0159] Network node 800 includes processing circuitry 802, memory 804, communication interface 806, and power supply 808. Network node 800 may include multiple physically separate components (e.g., node B components and RNC components, or BTS components and BSC components, etc.), each component may have its own corresponding components. In a specific scenario where network node 800 includes multiple separate components (e.g., BTS and BSC components), one or more separate components may be shared among multiple network nodes. For example, a single RNC can control multiple node Bs. In such scenarios, under certain circumstances, each unique node B and RNC pair may be considered a single, separate network node. In some embodiments, network node 800 may be configured to support multiple RATs. In such embodiments, some components may be duplicated (e.g., separate memory 804 for different RATs), while some components may be reused (e.g., the same antenna 810 may be shared by different RATs). Network node 800 may also include a variety of example components for integrating different wireless technologies (e.g., GSM, WCDMA, LTE, NR, Wi-Fi, Zigbee, Z-wave, LoRaWAN, RFID, or Bluetooth wireless technologies) into network node 800. These wireless technologies may be integrated into the same or different chips or chipsets and other components within network node 800.

[0160] Processing circuitry 802 may include one or more of the following, operable to provide network node 800 functionality individually or in combination with other network node 800 components (such as memory 804): microprocessor, controller, microcontroller, CPU, DSP, ASIC, FPGA, or any other suitable computing device, resource, or combination of hardware, software and / or coding logic.

[0161] In some embodiments, the processing circuitry 802 includes a system-on-a-chip (SOC). In some embodiments, the processing circuitry 802 includes one or more of a radio frequency (RF) transceiver circuitry 812 and a baseband processing circuitry 814. In some embodiments, the RF transceiver circuitry 812 and the baseband processing circuitry 814 may be on separate chips (or chipsets), boards, or units (such as radio units and digital units). In alternative embodiments, some or all of the RF transceiver circuitry 812 and the baseband processing circuitry 814 may be on the same chip or chipset, board, or unit.

[0162] Memory 804 may include any form of volatile or non-volatile computer-readable memory, including but not limited to permanent storage devices, solid-state memory, remotely mounted memory, magnetic media, optical media, RAM, ROM, mass storage media (e.g., hard disk), removable storage media (e.g., flash drives, optical discs (CDs), or digital video discs (DVDs)), and / or any other volatile or non-volatile, non-transitory device-readable and / or computer-executable storage device that stores information, data, and / or instructions that can be used by processing circuitry 802. Memory 804 may store any suitable instructions, data, or information, including computer programs, software, applications (including one or more of logic, rules, code, tables, etc.), and / or other instructions that can be executed by processing circuitry 802 and utilized by network node 800. Memory 804 may be used to store any calculations performed by processing circuitry 802 and / or any data received via communication interface 806. In some embodiments, processing circuitry 802 and memory 804 are integrated.

[0163] Communication interface 806 is used in wired or wireless communication of signaling and / or data between network nodes, access networks, and / or UEs. As shown, communication interface 806 includes a port / terminal 816 for sending and receiving data to and from the network, for example, via a wired connection. Communication interface 806 also includes radio front-end circuitry 818 that may be coupled to antenna 810 or, in a particular embodiment, is part of antenna 810. Radio front-end circuitry 818 includes a filter 820 and an amplifier 822. Radio front-end circuitry 818 may be connected to antenna 810 and processing circuitry 802. Radio front-end circuitry 818 may be configured to modulate the signal transmitted between antenna 810 and processing circuitry 802. Radio front-end circuitry 818 may receive digital data that will be transmitted to other network nodes or UEs via a wireless connection. Radio front-end circuitry 818 may use a combination of filter 820 and / or amplifier 822 to convert the digital data into radio signals with suitable channel and bandwidth parameters. The radio signals may then be transmitted via antenna 810. Similarly, when receiving data, antenna 810 can collect radio signals, which are then converted into digital data by radio front-end circuitry 818. The digital data can then be passed to processing circuitry 802. In other embodiments, the communication interface may include different components and / or different combinations of components.

[0164] In certain alternative embodiments, network node 800 does not include a separate radio front-end circuitry 818; instead, processing circuitry 802 includes radio front-end circuitry and is connected to antenna 810. Similarly, in some embodiments, all or part of RF transceiver circuitry 812 is part of communication interface 806. In other embodiments, communication interface 806 includes one or more ports or terminals 816, radio front-end circuitry 818, and RF transceiver circuitry 812 as part of a radio unit (not shown), and communication interface 806 communicates with baseband processing circuitry 814, which is part of a digital unit (not shown).

[0165] Antenna 810 may include one or more antennas or antenna arrays configured to transmit and / or receive wireless signals. Antenna 810 may be coupled to radio front-end circuitry 818 and may be any type of antenna capable of wirelessly transmitting and receiving data and / or signals. In a particular embodiment, antenna 810 is decoupled from network node 800 and may be connected to network node 800 via an interface or port.

[0166] Antenna 810, communication interface 806, and / or processing circuitry 802 can be configured to perform any receive operation and / or specific acquisition operation described herein as being performed by a network node. Any information, data, and / or signals can be received from the UE, another network node, and / or any other network device. Similarly, antenna 810, communication interface 806, and / or processing circuitry 802 can be configured to perform any transmit operation described herein as being performed by a network node. Any information, data, and / or signals can be transmitted to the UE, another network node, and / or any other network device.

[0167] Power supply 808 provides power to the various components of network node 800 in a form suitable for the respective components (e.g., at the voltage and current levels required by each respective component). Power supply 808 may also include or be coupled to power management circuitry to provide power to the components of network node 800 for performing the functions described herein. For example, network node 800 may be connected to an external power source (e.g., mains, power outlet) via input circuitry or an interface (such as a cable), whereby the external power source provides power to the power circuitry of power supply 808. As yet another example, power supply 808 may include a power source in the form of a battery or battery pack connected to or integrated into the power circuitry. The battery can provide backup power if the external power source fails.

[0168] Embodiments of network node 800 may include, in addition to Figure 8Additional components beyond those shown, which provide specific aspects of the functionality of the network node, including any functionality described herein and / or any functionality necessary to support the topics described herein. For example, network node 800 may include a user interface device to allow information to be input into and output from network node 800. This allows users to perform diagnostic, maintenance, repair, and other management functions on network node 800. In some embodiments, providing a core network node (such as...) may be omitted. Figure 6 The core network node 108), and some components (such as radio front-end circuit 815 and RF transceiver circuit 812).

[0169] Figure 9 This is a block diagram illustrating a virtualized environment 900 in which functionality implemented by some embodiments can be virtualized. In the current context, virtualization means creating virtual versions of devices or equipment, which may include virtualized hardware platforms, storage devices, and networking resources. As used herein, virtualization can be applied to any device or component thereof described herein and relates to an implementation in which at least a portion of functionality is implemented as one or more virtual components. Some or all of the functionality described herein can be implemented as virtual components executed by one or more virtual machines (VMs) in one or more virtual environments 900 hosted by one or more hardware nodes, such as hardware computing devices operating as network nodes, UEs, core network nodes, or hosts. Furthermore, in embodiments in which virtual nodes do not require radio connectivity (e.g., core network nodes or hosts), the nodes can be fully virtualized. In some embodiments, the virtualized environment 900 includes components defined by the O-RAN Alliance, such as an O-Cloud environment orchestrated by a service management and orchestration framework via an O-2 interface. Virtualization can facilitate the distributed implementation of network nodes, UEs, core network nodes, or hosts.

[0170] Application 902 (which may alternatively be referred to as a software instance, virtual device, network function, virtual node, virtual network function, etc.) runs in virtualization environment 900 to implement certain features, functions, and / or benefits of some embodiments disclosed herein.

[0171] Hardware 904 includes processing circuitry, memory storing software and / or instructions executable by the hardware processing circuitry, and / or other hardware devices as described herein, such as network interfaces, input / output interfaces, etc. Software may be executed by the processing circuitry to instantiate one or more virtualization layers 906 (also referred to as a hypervisor or virtual machine monitor (VMM)), provide virtual machines 908A and 908B (one or more of which may generally be referred to as virtual machine 908), and / or perform any functionality, features, and / or benefits described for some embodiments described herein. Virtualization layer 906 may present a virtual operating platform to virtual machine 908 that appears to be networked hardware.

[0172] Virtual machine 908 includes virtual processing, virtual memory, virtual network or interface, and virtual storage devices, and can be run by a corresponding virtualization layer 906. Different embodiments of instances of virtual device 902 can be implemented on one or more virtual machines 908, and can be implemented in different ways. In some contexts, hardware virtualization is referred to as Network Functions Virtualization (NFV). NFV can be used to consolidate many types of network devices into industry-standard, high-capacity server hardware, physical switches, and physical storage devices that can reside in data centers and client devices.

[0173] In the context of NFV, a virtual machine 908 can be a software implementation of a physical machine, which runs programs as if they were executed on a physical, non-virtualized machine. Each virtual machine 908, along with the portion of hardware 904 that executes that virtual machine (hardware dedicated to that virtual machine and / or hardware shared by that virtual machine with other virtual machines), forms a separate virtual network unit. Still within the context of NFV, the virtual network function is responsible for handling specific network functions running on one or more virtual machines 908 above hardware 904, and corresponds to application 902.

[0174] Hardware 904 can be implemented in a standalone network node with general or specific components. Hardware 904 may implement some functions via virtualization. Alternatively, hardware 904 may be part of a larger hardware cluster (e.g., in a data center or CPE) where many hardware nodes work together and are managed via management and orchestration 910, which, among other things, oversees the lifecycle management of application 902. In some embodiments, hardware 904 is coupled to one or more radio units, including one or more transmitters and one or more receivers that can be coupled to one or more antennas. The radio units may communicate directly with other hardware nodes via one or more suitable network interfaces and may be used in combination with virtual components to provide a radio-capable virtual node, such as a radio access node or base station. In some embodiments, a control system 912 may be used to provide signaling, which may alternatively be used for communication between the hardware nodes and the radio units.

[0175] While the computing devices described herein (e.g., UE, network node) may include combinations of the hardware components shown, other embodiments may include computing devices with different combinations of components. It will be understood that these computing devices may include any suitable combination of hardware and / or software required to perform the tasks, features, functions, and methods disclosed herein. The determination, calculation, acquisition, or similar operations described herein may be performed by processing circuitry that may process information, for example, by: converting acquired information into other information, comparing the acquired or converted information with information stored in a network node, and / or performing one or more operations based on the acquired or converted information, and making a determination as a result of said processing. Furthermore, although components are depicted as single boxes located within larger boxes or nested within multiple boxes, in practice, a computing device may include multiple different physical components constituting a single illustrated component, and functionality may be partitioned between individual components. For example, a communication interface may be configured to include any of the components described herein, and / or the functionality of a component may be partitioned between processing circuitry and the communication interface. In another example, non-computationally intensive functions of any such component may be implemented in software or firmware, and computationally intensive functions may be implemented in hardware.

[0176] In certain embodiments, some or all of the functions described herein may be provided by processing circuitry executing instructions stored in memory, which may be a computer program product in the form of a non-transitory computer-readable storage medium. In alternative embodiments, some or all of the functions may be provided by processing circuitry without requiring, for example, hard-wired execution of instructions stored on a separate or separate device-readable storage medium. In any of these particular embodiments, processing circuitry may be configured to perform the described functions regardless of whether instructions stored on a non-transitory computer-readable storage medium are executed. The benefits provided by such functions are not limited to the processing circuitry or other components of the computing device, but are enjoyed by the computing device as a whole and / or generally by the end user and wireless network.

[0177] The following describes some of the references referred to in this article as the background art: 1. Radio Access Networks Technical Specification Group; Study on Artificial Intelligence (AI) / Machine Learning (ML) for NR Air Interface (Revision 18), 3GPP TR 38.843, December 2023; 2. "Wireless Channel Charting: Theory, Practice, and Applications" by Ferrand, Paul, Maxime Guillaud, Christoph Studer, and Olav Tirkkonen. IEEE Communications Magazine Volume 61, Issue 6 (2023): pp. 124-130; 3. NG Radio Access Network (NG-RAN); Phase 2 functional specification for User Equipment (UE) location in NG-RAN, 3GPP TR 38.305, December 2023.

[0178] Those skilled in the art will recognize improvements and modifications to the embodiments of this disclosure. All such improvements and modifications are considered to be within the scope of the concepts disclosed herein.

[0179] Some embodiments of this disclosure include: Example 1: A method performed by a first network node (502) for facilitating the collection and use of location data as training data for training an artificial intelligence / machine learning (AI / ML) model, the method comprising: determining (512) a tolerance for ground truth labeling based on one or more Quality of Service (QoS) criteria; determining (514) one or more reliability criteria for the location data based on the tolerance for ground truth labeling; providing (516) the one or more reliability criteria for the location data to a second network node (504); and receiving (522) the collected location data from the second network node (504).

[0180] Example 2: According to the method described in Example 1, the first network node (502) is at least one of a base station device, a user equipment device (UE), or a location management function (LMF), and the second network node (504) is at least one of a base station device, a UE, or an LMF, and the first network node (502) and the second network node (504) are of different types.

[0181] Example 3: The method according to any one of Examples 1 to 2, wherein the QoS standard is received from a client application or from another network node.

[0182] Example 4: The method according to any one of Examples 1 to 3, wherein the one or more reliability criteria are determined based on a predefined configuration at one of the first network node (502) or the second network node (504).

[0183] Example 5: The method according to any one of Examples 1 to 4 further includes: receiving (510) capability information from the second network node (504), wherein the one or more reliability criteria are at least partially based on the capability information.

[0184] Example 6: The method according to any one of Examples 1 to 5, wherein the one or more reliability criteria define limits associated with alarm limits, maximum positioning error, uncertainty level, and confidence level of the positioning data.

[0185] Example 7: The method according to Example 6, wherein the alarm limit, the positioning error, the uncertainty level and the confidence level are based on a geographical region description associated with the positioning data.

[0186] Example 8: The method according to any one of Examples 1 to 7, wherein the one or more reliability criteria are based on measurement results obtained from uplink measurements from a UE (506) with a known location.

[0187] Example 9: The method according to any one of Examples 1 to 7, wherein the one or more reliability criteria are based on measurement results obtained from downlink measurements at a UE with a known location.

[0188] Example 10: The method according to any one of Examples 8 to 9, wherein the UE is a Positioning Reference Unit (PRU).

[0189] Example 11: According to the method of Example 9, the second network node (504) obtains the measurement result from the UE having the known location, wherein the difference between the measurement result from the UE and the uplink measurement result from the second network node (504) is less than a predefined threshold.

[0190] Example 12: The method according to any one of Examples 1 to 11, wherein the collected positioning data includes measurement data and label data, wherein the measurement data of the collected positioning data meets a first minimum quality standard, and wherein, in response to the label data not meeting a second minimum quality standard, the collected positioning data is labeled as unlabeled training data, and in response to the label data meeting the second minimum quality standard, the collected positioning data is labeled as labeled training data.

[0191] Example 13: The method according to Example 12 further includes: checking (524) the quality of the measurement data in response to the occurrence of one or more of the following: transmitter implementation error; receiver implementation error; radio channel quality between the transmitter and receiver not meeting a predefined threshold; or a predefined radio signal configuration.

[0192] Example 14: A first network node (502) for determining whether location data should be collected and used as training data for training an artificial intelligence / machine learning AI / ML model, including processing circuitry for performing any of the methods described in Examples 1 to 13.

[0193] Example 15: A method performed by a second network node (504) for determining whether location data should be collected and used as training data for training an artificial intelligence / machine learning (AI / ML) model, the method comprising: receiving (516) one or more reliability criteria for location data from a first network node (502); collecting (518) location data that satisfies the one or more reliability criteria to obtain the collected location data; and providing (522) the collected location data to the first network node (502).

[0194] Example 16: The method described in Example 15 further includes: providing (510) capability information to the first network node (502).

[0195] Example 17: The method according to any one of Examples 15 to 16, wherein the first network node (502) is at least one of a base station device, a user equipment device (UE) or a location management function (LMF), and wherein the second network node (504) is at least one of a base station device, a UE or an LMF, and wherein the first network node (502) and the second network node (504) are of different types.

[0196] Example 18: The method according to any one of Examples 15 to 17, wherein the one or more reliability criteria are determined based on a predefined configuration at one of the first network node (502) or the second network node (504).

[0197] Example 19: The method according to any one of Examples 15 to 18, wherein the one or more reliability criteria define limits associated with alarm limits, maximum positioning error, uncertainty level, and confidence level of the positioning data.

[0198] Example 20: The method according to Example 19, wherein the alarm limit, the maximum positioning error, the uncertainty level and the confidence level are based on a geographical region description associated with the positioning data.

[0199] Example 21: The method according to any one of Examples 15 to 20, wherein the one or more reliability criteria are based on measurement results obtained from uplink measurements from a UE with a known location.

[0200] Example 22: The method according to any one of Examples 15 to 20, wherein the one or more reliability criteria are based on measurement results obtained from downlink measurements at a UE (506) with a known location.

[0201] Example 23: The method according to any one of Examples 21 to 22, wherein the UE (506) is a positioning reference unit (PRU).

[0202] Example 24: The method according to any one of Examples 21 to 23 further includes: obtaining (520) the measurement result from the UE (506) having the known location, wherein the difference between the measurement result from the UE and the uplink measurement result from the second network node (504) is less than a predefined threshold.

[0203] Example 25: A second network node (504) for determining whether location data should be collected and used as training data for training an artificial intelligence / machine learning AI / ML model, including processing circuitry for performing any of the methods described in Examples 1 to 13.

Claims

1. A method performed by a second network node (504) for facilitating the collection and use of location data as training data for training artificial intelligence / machine learning AI / ML models, the method comprising: Receive (516) one or more reliability criteria for locating data from the first network node (502); as well as Collect (518) positioning data that meets one or more of the reliability criteria to obtain the collected positioning data.

2. The method according to claim 1, wherein, The one or more reliability criteria are associated with the confidence level of the location data.

3. The method according to any one of claims 1 to 2, wherein, The first network node (502) is at least one of a base station, a user equipment (UE) or a location management function (LMF), and the second network node (504) is at least one of a base station, a UE or an LMF, and the first network node (502) and the second network node (504) are of different types.

4. The method according to any one of claims 1 to 3, wherein, The one or more reliability criteria are determined based on a predefined configuration at either the first network node (502) or the second network node (504).

5. The method according to any one of claims 1 to 4, wherein, The one or more reliability criteria define limits associated with the location data, including alarm limits, maximum location error, uncertainty level, and confidence level.

6. The method according to claim 5, wherein, The alarm limit, the maximum positioning error, the uncertainty level, and the confidence level are based on a description of the geographic region associated with the positioning data.

7. The method according to any one of claims 1 to 6, wherein, The one or more reliability criteria are based on measurement results obtained from uplink measurements from UEs with known locations.

8. The method according to any one of claims 1 to 6, wherein, The one or more reliability criteria are based on measurements obtained from downlink measurements at a UE (506) with a known location.

9. The method according to any one of claims 7 to 8, wherein, The UE (506) is a positioning reference unit (PRU).

10. The method according to any one of claims 7 to 9, further comprising: The measurement result is obtained (520) from the UE (506) having the known location, wherein the difference between the measurement result from the UE and the uplink measurement result from the second network node (504) is less than a predefined threshold.

11. The method according to any one of claims 1 to 10, further comprising: Provide the location data collected by (522) to the first network node (502).

12. A second network node (504) for facilitating the collection and use of location data as training data for training artificial intelligence / machine learning AI / ML models, comprising processing circuitry that enables the second network node (504): Receive (516) one or more reliability criteria for locating data from the first network node (502); and Collect (518) positioning data that meets one or more of the reliability criteria to obtain the collected positioning data.

13. The second network node (504) according to claim 12, wherein, The one or more reliability criteria are quality indicators associated with the location data.

14. The second network node (504) according to any one of claims 12 to 13, wherein, The first network node (502) is at least one of a base station, a user equipment (UE) or a location management function (LMF), and the second network node (504) is at least one of a base station, a UE or an LMF, and the first network node (502) and the second network node (504) are of different types.

15. The second network node (504) according to any one of claims 12 to 14, wherein, The one or more reliability criteria are determined based on a predefined configuration at either the first network node (502) or the second network node (504).

16. The second network node (504) according to any one of claims 12 to 15, wherein, The one or more reliability criteria define limits associated with the location data, including alarm limits, maximum location error, uncertainty level, and confidence level.

17. The second network node (504) according to claim 16, wherein, The alarm limit, the maximum positioning error, the uncertainty level, and the confidence level are based on a description of the geographic region associated with the positioning data.

18. The second network node (504) according to any one of claims 12 to 17, wherein, The one or more reliability criteria are based on measurement results obtained from uplink measurements from UEs with known locations.

19. The second network node (504) according to any one of claims 12 to 17, wherein, The one or more reliability criteria are based on measurements obtained from downlink measurements at a UE (506) with a known location.

20. The second network node (504) according to any one of claims 18 to 19, wherein, The UE (506) is a positioning reference unit (PRU).

21. The second network node (504) according to any one of claims 18 to 20, wherein, The processing circuit further enables the second network node (504): The measurement result is obtained (520) from the UE (506) having the known location, wherein the difference between the measurement result from the UE and the uplink measurement result from the second network node (504) is less than a predefined threshold.

22. The second network node (504) according to any one of claims 12 to 21, wherein, The processing circuit further enables the second network node (504): Provide the location data collected by (522) to the first network node (502).

23. A method performed by a first network node (502) for facilitating the collection and use of location data as training data for training an artificial intelligence / machine learning AI / ML model, the method comprising: Based on one or more Quality of Service (QoS) standards, determine (512) the tolerance for real value labeling; Based on the tolerance used for true value labeling, determine (514) one or more reliability criteria for locating the data; Provide the second network node (504) with (516) the one or more reliability criteria for locating the data, and Receive the location data collected by (522) from the second network node (504).

24. The method according to claim 23, wherein, The first network node (502) is at least one of a base station, a user equipment (UE) or a location management function (LMF), and the second network node (504) is at least one of a base station, a UE or an LMF, and the first network node (502) and the second network node (504) are of different types.

25. The method according to any one of claims 23 to 24, wherein, The QoS standard is received from the client application or from another network node.

26. The method according to any one of claims 23 to 25, wherein, The one or more reliability criteria are determined based on a predefined configuration at either the first network node (502) or the second network node (504).

27. The method according to any one of claims 23 to 26, further comprising: (510) Capability information is received from the second network node (504), wherein the one or more reliability criteria are at least partially based on the capability information.

28. The method according to any one of claims 23 to 27, wherein, The one or more reliability criteria define limits associated with the location data, including alarm limits, maximum location error, uncertainty level, and confidence level.

29. The method according to claim 28, wherein, The alarm limit, the location error, the uncertainty level, and the confidence level are based on a description of the geographic region associated with the location data.

30. The method according to any one of claims 23 to 29, wherein, The one or more reliability criteria are based on measurements obtained from uplink measurements from a UE (506) with a known location.

31. The method according to any one of claims 23 to 29, wherein, The one or more reliability criteria are based on measurements obtained from downlink measurements at a UE with a known location.

32. The method according to any one of claims 30 to 31, wherein, The UE is a Positioning Reference Unit (PRU).

33. The method according to claim 31, wherein, The second network node (504) obtains the measurement result from the UE having the known location, wherein the difference between the measurement result from the UE and the uplink measurement result from the second network node (504) is less than a predefined threshold.

34. The method according to any one of claims 23 to 33, wherein, The collected location data includes measurement data and tag data, wherein the measurement data of the collected location data meets a first minimum quality standard, and wherein, In response to the label data failing to meet the second minimum quality standard, the collected location data is labeled as unlabeled training data. In response to the label data meeting the second minimum quality standard, the collected location data is labeled as labeled training data.

35. The method of claim 34, further comprising: In response to one or more of the following occurrences, check the quality of the measurement data described in (524): Transmitter error; Receiver implementation error; The radio channel quality between the transmitter and receiver does not meet a predefined threshold; or Predefined radio signal configuration.

36. A first network node (502) for facilitating the collection and use of location data as training data for training artificial intelligence / machine learning AI / ML models, comprising processing circuitry to: Based on one or more Quality of Service (QoS) standards, determine (512) the tolerance for real value labeling; Based on the tolerance used for true value labeling, determine (514) one or more reliability criteria for locating the data; Provide the second network node (504) with (516) the one or more reliability criteria for locating the data, and Receive the location data collected by (522) from the second network node (504).

37. The first network node (502) according to claim 36, wherein, The processing circuit is further configured to perform any of the methods according to claims 24 to 35.

Images