Machine learning positioning aspects for AIML models at network nodes

Abstract

The various example embodiments described herein relate to apparatuses and methods for machine learning localization. One such example embodiment relates to an apparatus comprising: at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus to at least: determine a localization configuration request based on the machine learning capabilities of a plurality of network nodes capable of performing localization-related measurements, the localization configuration request indicating at least one localization configuration, the at least one localization configuration including at least a localization method to be applied to estimate the location of a user equipment; send the localization configuration request to at least one of the plurality of network nodes; receive at least one response from the at least one of the plurality of network nodes, wherein the at least one response includes at least one selected localization configuration; and configure the at least one localization configuration to the at least one of the plurality of network nodes based on the at least one response received from the at least one of the plurality of network nodes.

Description

Technical Field

[0001] This invention relates to apparatus, methods, and computer program products for ML (machine learning) localization. Background Technology

[0002] The following meanings apply to the abbreviations used in this specification:

[0003] AI (Artificial Intelligence)

[0004] AIML (Artificial Intelligence and Machine Learning)

[0005] AMF Access and Mobility Functions

[0006] BW bandwidth

[0007] CHO condition switching

[0008] CIR channel impulse response

[0009] CNN (Convolutional Neural Network)

[0010] CSI Channel State Information

[0011] DL downlink

[0012] DP Latency Profile

[0013] HO switch

[0014] LCM – Lifecycle Management

[0015] LCS Location Service

[0016] LMF location management function

[0017] LPP LTE positioning protocol

[0018] ML Machine Learning

[0019] NF Network Functions

[0020] NRPPa NR Positioning Protocol A

[0021] NW Network

[0022] PDP power delay distribution

[0023] PRS Positioning Reference Signal

[0024] ResNet residual neural network

[0025] RNN (Recurrent Neural Network)

[0026] SRS Detection Reference Signal

[0027] ToA Arrival Time

[0028] TRP Transmit and Receive Points

[0029] TX transmission

[0030] UE – User Equipment

[0031] UL uplink

[0032] UPF User Face Functions

[0033] The example implementation (though not limited to) relates to the UE localization process. UE localization can be performed using AI / ML technologies. However, some issues may arise when applying AI / ML technologies. Summary of the Invention

[0034] The example implementation addresses this situation and aims to provide improved UE positioning processing.

[0035] Several aspects of the various example embodiments will be described with reference to certain aspects. These aspects are not intended to indicate key or essential features of the various example embodiments, nor are they intended to be used to otherwise limit the scope of this disclosure. Other features, aspects, and elements of the various example embodiments will be readily apparent to those skilled in the art in light of this disclosure.

[0036] According to a first aspect, an apparatus is provided, comprising: At least one processor, and At least one memory storing instructions that, when executed by at least one processor, cause the device to at least: Based on the machine learning capabilities of multiple network nodes capable of performing location-related measurements, a location configuration request is determined, the location configuration request indicating at least one location configuration, the at least one location configuration including at least a location method to be applied to estimate the location of the user equipment; Send a location configuration request to at least one of multiple network nodes; Receive at least one response from at least one of a plurality of network nodes, wherein the at least one response includes at least one selected location configuration; and Based on at least one response received from at least one of the multiple network nodes, at least one location configuration is configured to at least one of the multiple network nodes.

[0037] The first aspect can be modified as follows:

[0038] Instructions stored in at least one memory, when executed by at least one processor, may cause the device to at least: The machine learning capabilities of network nodes across multiple network nodes are determined as follows: Request information from network nodes about their machine learning capabilities, and Receive a response from a network node, which includes information about machine learning-related capabilities.

[0039] Machine learning capabilities can include information about the machine learning models and / or machine learning functions applied by each of the multiple network nodes.

[0040] Information about machine learning models and / or machine learning functions may include: the type, characteristics, features, and / or structure of the machine learning models and / or machine learning functions, and / or Inputs and / or outputs of machine learning models and / or machine learning functions, and / or Information regarding the computational complexity and / or interference latency of machine learning models and / or machine learning functions.

[0041] The output of a machine learning model and / or machine learning function may include at least one supported intermediate feature.

[0042] The inputs to machine learning models and / or machine learning functions may include at least one of the following: Supported uplink probe reference signal configurations, Supported channel pulse formats, Supported power delay distribution formats, and / or Supported delay distribution formats.

[0043] The location configuration indicated by the location configuration request may further include at least one of the following: The machine learning model and / or machine learning function to be applied. The inputs and / or outputs of machine learning models and / or machine learning functions, and Uplink detection reference signal configuration.

[0044] Instructions stored in at least one memory, when executed by at least one processor, may cause the device to at least: Based on the machine learning capabilities of multiple network nodes, the location configuration request to the multiple network nodes is determined as follows: Alignment will be achieved by machine learning models and / or machine learning functions applied by multiple network nodes; The alignment will be applied to the format of the intermediate features of the aligned machine learning model; and Align uplink probe reference signal configuration.

[0045] Instructions stored in at least one memory, when executed by at least one processor, may cause the device to at least: Based on the responses received from all network nodes across multiple network nodes, the location configuration is configured for at least one of the multiple network nodes as follows: Determine which location configuration or which location configurations are supported by all network nodes across multiple network nodes.

[0046] Instructions stored in at least one memory, when executed by at least one processor, may cause the device to at least: Configure all network nodes of multiple network nodes to have the same location configuration, or Configure each network node individually with a location configuration.

[0047] Instructions stored in at least one memory, when executed by at least one processor, may cause the device to at least: Receive location-related measurement results from multiple network nodes, and The location of the user equipment is determined based on the combination of received positioning-related measurements.

[0048] Instructions stored in at least one memory, when executed by at least one processor, at least: Intermediate features are combined by applying weights to the positioning-related measurement results.

[0049] Instructions stored in at least one memory, when executed by at least one processor, may also cause the device to: The weights of intermediate features received from multiple network nodes are adjusted based on the user equipment's movement.

[0050] The location-related measurement results may include intermediate features.

[0051] According to a second aspect, an apparatus is provided, comprising: At least one processor, and At least one memory storing instructions that, when executed by at least one processor, cause the device to at least: The network control element receives a location configuration request, which indicates at least one location configuration, the at least one location configuration including at least a location method to be applied to estimate the location of the user equipment; Based on the machine learning capabilities of the device, at least one location configuration of the location configuration request is selected. Send a response to the network control element including at least one selected positioning configuration; Receive information about the location configuration to be applied; and Based on the location configuration to be applied, location-related measurements are performed using a machine learning model.

[0052] The second aspect can be modified as follows:

[0053] Instructions stored in at least one memory, when executed by at least one processor, may cause the device to at least: Receive requests from network control elements for information regarding the machine learning capabilities of the device; and Send a response to the network control element, including information about machine learning-related capabilities.

[0054] The machine learning-related capabilities may include information about the machine learning models and / or machine learning functions applied by the device.

[0055] Information about the machine learning model and / or machine learning functionality may include: The type, characteristics, features, and / or structure of machine learning models and / or machine learning functions, and / or The input and / or output of a machine learning model, and / or Information regarding the computational complexity of machine learning models and / or the interference latency of machine learning models and / or machine learning functions.

[0056] The output of a machine learning model and / or machine learning function may include at least one supported intermediate feature.

[0057] The output of machine learning models and / or machine learning functions may include at least one of the following Supported uplink probe reference signal configurations, Supported channel pulse formats, Supported power delay distribution formats, and / or Supported delay distribution formats.

[0058] The location configuration indicated by the location configuration request may further include at least one of the following: The machine learning model and / or machine learning function to be applied; The inputs and / or outputs of machine learning models and / or machine learning functions; and Uplink detection reference signal configuration.

[0059] Instructions stored in at least one memory, when executed by at least one processor, may cause the device to at least: Send the results of positioning-related measurements to the network control element.

[0060] This result can be an intermediate feature.

[0061] According to the third aspect, a method is provided, the method comprising: Based on the machine learning capabilities of multiple network nodes capable of performing location-related measurements, a location configuration request is determined, the location configuration request indicating at least one location configuration, the at least one location configuration including at least a location method to be applied to estimate the location of the user equipment; Send a location configuration request to at least one of multiple network nodes; Receive at least one response from at least one of a plurality of network nodes, wherein the at least one response includes at least one selected location configuration; and Based on at least one response received from at least one of the multiple network nodes, at least one location configuration is configured to at least one of the multiple network nodes.

[0062] The third aspect can be modified as follows:

[0063] The method may also include: The machine learning capabilities of network nodes across multiple network nodes are determined as follows: Request information from network nodes about their machine learning capabilities, and Receive a response from a network node, which includes information about machine learning-related capabilities.

[0064] Machine learning capabilities can include information about the machine learning models and / or machine learning functions applied by each of the multiple network nodes.

[0065] Information about machine learning models and / or machine learning functions may include: the type, characteristics, features, and / or structure of the machine learning models and / or machine learning functions, and / or Inputs and / or outputs of machine learning models and / or machine learning functions, and / or Information regarding the computational complexity and / or interference latency of machine learning models and / or machine learning functions.

[0066] The output of a machine learning model and / or machine learning function may include at least one supported intermediate feature.

[0067] The inputs to machine learning models and / or machine learning functions may include at least one of the following: Supported uplink probe reference signal configurations, Supported channel pulse formats, Supported power delay distribution formats, and / or Supported delay distribution formats.

[0068] The location configuration indicated by the location configuration request may further include at least one of the following: The machine learning model and / or machine learning function to be applied. The inputs and / or outputs of machine learning models and / or machine learning functions, and Uplink detection reference signal configuration.

[0069] The method may also include: Based on the machine learning capabilities of multiple network nodes, the location configuration request to the multiple network nodes is determined as follows: Alignment will be achieved by machine learning models and / or machine learning functions applied by multiple network nodes; The alignment will be applied to the format of the intermediate features of the aligned machine learning model; and Align uplink probe reference signal configuration.

[0070] The method may also include: Based on the responses received from all network nodes across multiple network nodes, the location configuration is configured for at least one of the multiple network nodes as follows: Determine which location configuration or which location configurations are supported by all network nodes across multiple network nodes.

[0071] The method may also include: Configure all network nodes of multiple network nodes to have the same location configuration, or Configure each network node individually with a location configuration.

[0072] The method may also include: Receive location-related measurement results from multiple network nodes, and The location of the user equipment is determined based on the combination of received positioning-related measurements.

[0073] The method may also include: Intermediate features are combined by applying weights to the positioning-related measurement results.

[0074] The method may also include: The weights of intermediate features received from multiple network nodes are adjusted based on the user equipment's movement.

[0075] The location-related measurement results may include intermediate features.

[0076] According to the fourth aspect, a method is provided in a network node, the method comprising: The network control element receives a location configuration request, which indicates at least one location configuration, the at least one location configuration including at least a location method to be applied to estimate the location of the user equipment; Based on the machine learning capabilities of the device, at least one location configuration of the location configuration request is selected. Send a response to the network control element including at least one selected positioning configuration; Receive information about the location configuration to be applied; and Based on the location configuration to be applied, location-related measurements are performed using a machine learning model.

[0077] The fourth aspect can be modified as follows:

[0078] The method may also include: Receive requests from network control elements for information regarding the machine learning capabilities of the device; and Send a response to the network control element, including information about machine learning-related capabilities.

[0079] The machine learning-related capabilities may include information about the machine learning models and / or machine learning functions applied by the device.

[0080] Information about the machine learning model and / or machine learning functionality may include: The type, characteristics, features, and / or structure of machine learning models and / or machine learning functions, and / or The input and / or output of a machine learning model, and / or Information regarding the computational complexity of machine learning models and / or the interference latency of machine learning models and / or machine learning functions.

[0081] The output of a machine learning model and / or machine learning function may include at least one supported intermediate feature.

[0082] The output of machine learning models and / or machine learning functions may include at least one of the following Supported uplink probe reference signal configurations, Supported channel pulse formats, Supported power delay distribution formats, and / or Supported delay distribution formats.

[0083] The location configuration indicated by the location configuration request may further include at least one of the following: The machine learning model and / or machine learning function to be applied; The inputs and / or outputs of machine learning models and / or machine learning functions; and Uplink detection reference signal configuration.

[0084] The method may also include: Send the results of positioning-related measurements to the network control element.

[0085] This result can be an intermediate feature.

[0086] According to a fifth aspect of the invention, a computer program product is provided, comprising code components (or instruction sets) for executing methods according to the third or fourth aspect and / or modifications thereof when run on a processing component or module. The computer program product may be embodied on a computer-readable medium, and / or the computer program product may be directly loaded into the internal memory of a computer, and / or may be transmitted via a network through at least one of the processes of uploading, downloading, and pushing.

[0087] According to a sixth aspect, an apparatus is provided, comprising: A component for determining a location configuration request based on the machine learning capabilities of multiple network nodes capable of performing location-related measurements, the location configuration request indicating at least one location configuration, the at least one location configuration including at least a location method to be applied to estimate the location of a user equipment. A component for sending a location configuration request to at least one of a plurality of network nodes; Components for receiving at least one response from at least one of a plurality of network nodes, wherein the at least one response includes at least one selected positioning configuration; and A component for configuring at least one positioning configuration to at least one of a plurality of network nodes based on at least one response received from at least one of a plurality of network nodes.

[0088] The sixth aspect can be modified as follows:

[0089] The device may also include: Components used to determine the machine learning-related capabilities of network nodes among multiple network nodes in the following way: Request information from network nodes about their machine learning capabilities, and Receive a response from a network node, which includes information about machine learning-related capabilities.

[0090] Machine learning capabilities can include information about the machine learning models and / or machine learning functions applied by each of the multiple network nodes.

[0091] Information about machine learning models and / or machine learning functions may include: the type, characteristics, features, and / or structure of the machine learning models and / or machine learning functions, and / or Inputs and / or outputs of machine learning models and / or machine learning functions, and / or Information regarding the computational complexity and / or interference latency of machine learning models and / or machine learning functions.

[0092] The output of a machine learning model and / or machine learning function may include at least one supported intermediate feature.

[0093] The inputs to machine learning models and / or machine learning functions may include at least one of the following: Supported uplink probe reference signal configurations, Supported channel pulse formats, Supported power delay distribution formats, and / or Supported delay distribution formats.

[0094] The location configuration indicated by the location configuration request may further include at least one of the following: The machine learning model and / or machine learning function to be applied. The inputs and / or outputs of machine learning models and / or machine learning functions, and Uplink detection reference signal configuration.

[0095] The device may also include: The components for machine learning-related capabilities based on multiple network nodes are determined by the following: (The components are then used to request location configuration from multiple network nodes.) Alignment will be achieved by machine learning models and / or machine learning functions applied by multiple network nodes; The alignment will be applied to the format of the intermediate features of the aligned machine learning model; and Align uplink probe reference signal configuration.

[0096] The device may also include: The component used to configure the location of at least one of the multiple network nodes based on responses received from all network nodes: Determine which location configuration or which location configurations are supported by all network nodes across multiple network nodes.

[0097] The device may also include: A component used to configure all network nodes of multiple network nodes to have the same positioning configuration, or A component used to individually configure network nodes with location settings for multiple network nodes.

[0098] The device may also include: Components for receiving positioning-related measurement results from multiple network nodes, and A component used to determine the location of a user equipment based on a combination of received location-related measurements.

[0099] The device may also include: A component used to combine intermediate features by applying weights to positioning-related measurements.

[0100] The device may also include: A component for adjusting the weights of intermediate features received from multiple network nodes based on the movement of user equipment.

[0101] The location-related measurement results may include intermediate features.

[0102] According to a seventh aspect, an apparatus is provided, comprising: A component for receiving a location configuration request from a network control element, the location configuration request indicating at least one location configuration, the at least one location configuration including at least a location method to be applied to estimate the location of a user equipment; A component for selecting at least one location configuration request based on the machine learning capabilities of the device; A component for sending a response to a network control element, including at least one selected positioning configuration; A component for receiving information about the positioning configuration to be applied; and A component used to perform location-related measurements using a machine learning model based on the location configuration to be applied.

[0103] The seventh aspect can be modified as follows:

[0104] The device may also include: A component for receiving a request from a network control element for information regarding the machine learning capabilities of the device; and A component used to send responses, including information about machine learning-related capabilities, to network control elements.

[0105] The machine learning-related capabilities may include information about the machine learning models and / or machine learning functions applied by the device.

[0106] Information about the machine learning model and / or machine learning functionality may include: The type, characteristics, features, and / or structure of machine learning models and / or machine learning functions, and / or The input and / or output of a machine learning model, and / or Information regarding the computational complexity of machine learning models and / or the interference latency of machine learning models and / or machine learning functions.

[0107] The output of a machine learning model and / or machine learning function may include at least one supported intermediate feature.

[0108] The output of machine learning models and / or machine learning functions may include at least one of the following Supported uplink probe reference signal configurations, Supported channel pulse formats, Supported power delay distribution formats, and / or Supported delay distribution formats.

[0109] The location configuration indicated by the location configuration request may further include at least one of the following: The machine learning model and / or machine learning function to be applied; The inputs and / or outputs of machine learning models and / or machine learning functions; and Uplink detection reference signal configuration.

[0110] The method may also include: Send the results of positioning-related measurements to the network control element.

[0111] This result can be an intermediate feature.

[0112] In all of the foregoing aspects and their various modifications, the intermediate feature may include information that can be used by network control elements to determine the location of user equipment.

[0113] In all of the above aspects and their various modifications, the localization method can be a machine learning-based localization method or a non-machine learning-based localization method. Attached Figure Description

[0114] These and other objectives, features, details, and advantages will become more apparent from the following detailed description of exemplary embodiments, taken in conjunction with the accompanying drawings, in which:

[0115] Figure 1A An LMF 1 according to an example embodiment is shown;

[0116] Figure 1B The process performed by LMF 1 according to the example embodiment is shown;

[0117] Figure 2A Node 2 is shown according to an example embodiment;

[0118] Figure 2BThe process performed by node 2 according to the example embodiment is shown;

[0119] Figure 3 The signal flow used for condition switching is shown;

[0120] Figures 4A to 4B Different AI / ML localization use cases are shown;

[0121] Figure 5 This demonstrates the features enabling ML, particularly the use of functional and model IDs with applicable condition information; and

[0122] Figure 6 (by) Figure 6A and Figure 6B The composition illustrates exemplary steps for case 3a according to an example embodiment. Detailed Implementation

[0123] In the following description, exemplary embodiments will be used. However, it should be understood that these descriptions are given by way of example only, and the described exemplary embodiments should in no way be construed as limiting the invention.

[0124] Before describing the example embodiments, the technical context of the example embodiments and the problems of the prior art are discussed in more detail below.

[0125] Some example implementations relate to ML positioning use cases. For instance, UE positioning is required for Conditional Handover (CHO), which occurs when... Figure 3 It is shown in the middle.

[0126] The first process, B1-B9, is similar to the baseline handover in NR Rel.15 [TS 38.300]. The configured event triggers the UE to send a measurement report. Based on this report, the source node can prepare one or more target cells for handover (CHO request + CHO request confirmation), and then send an RRC reconfiguration (CHO command) to the UE.

[0127] For NR Rel.15 baseline handover, the UE will immediately access the target cell to complete the handover. Conversely, for CHO, the UE will only access the target cell after the additional CHO execution conditions expire (i.e., the HO preparation and execution phases are decoupled). This condition is configured by the source node in the HO command.

[0128] Once the UE completes the handover to the target cell (e.g., the UE has sent an RRC reconfiguration completion message), the target cell sends a "Handover Successful" indication to the source cell. Upon receiving this indication from the target cell, the source cell stops its TX / RX to / from the UE and begins forwarding data to the target cell in step 18. Additionally, upon receiving the "HO Successful" indication, the source cell can release any (no longer needed) CHO preparations in other target nodes / cells.

[0129] The advantage of CHO is that the HO command can be sent very early while the UE is still secure in the source cell, without jeopardizing access in the target cell or the stability of its radio link. Conditional handover provides mobility robustness.

[0130] Additionally, some example implementations relate to Rel-18 research project (SI) [3GPP RP-213599] concerning artificial intelligence (AI) / machine learning (ML) for NR air interfaces.

[0131] This SI aims to explore the advantages of enhancing the air interface by leveraging features of AI / ML-based algorithms that enable improved performance and / or reduced complexity / overhead. The goal of this SI is to lay the foundation for future air interface use cases utilizing AI / ML technologies. The initial set of use cases to be covered includes CSI feedback enhancements (e.g., overhead reduction, improved accuracy, prediction), beam management (e.g., beam prediction in the temporal and / or spatial domains for overhead and latency reduction, improved beam selection accuracy), and positioning accuracy enhancements. For these use cases, their advantages should be assessed (using developed methodologies and defined KPIs), and their potential impact on specifications should be evaluated, including PHY layer aspects and protocol aspects. This SI commenced at the RAN1 / RAN2 meeting in May within 3GPP.

[0132] One of the key expected outcomes of this SI is that "the AI / ML approaches used for the selected sub-use cases need to be sufficiently diverse to support a wide range of needs regarding the gNB-UE collaboration layer."

[0133] It is important to note that in the WI phase of "AI / ML for Air Interface," additional use cases may be addressed. Starting with version 18, companies are likely to propose numerous use cases and applications of ML in gNB and UE. The goal is to explore ways to enhance the air interface by leveraging features that enable improved support for AI / ML-based algorithms to enhance performance and / or reduce complexity / overhead. The performance enhancements here depend on the use cases considered and could be, for example, improved throughput, robustness, accuracy, or reliability. The goal is to identify enough use cases to enable the identification of public AI / ML frameworks, including the functional requirements of the AI / ML architecture, which can be used in subsequent projects. The study should also identify areas where AI / ML can improve the performance of air interface functions. Specification impacts will be evaluated to improve the overall understanding of what is needed to enable AI / ML technologies for the air interface.

[0134] In the following text, some AI / ML localization use cases are described.

[0135] Figure 4A Case 1 is illustrated – UE-based localization using a UE-side model: (a) direct AI / ML localization and (b) AI / ML-assisted localization. In this case, the model is deployed on the UE side. Here, the direct AI / ML and AI / ML-assisted localization sub-use cases can be deployed accordingly to obtain output horizontal position or intermediate features. For both cases, the parameter measurements used as input are performed on the UE (downlink localization), and the final position estimation is also on the UE side. Illustrations indicating the reference signals used to generate the measurements and the entities involved in each case are shown in [the diagram]. Figure 4A It is presented in the middle.

[0136] Figure 4B Case 2a is illustrated—UE-assisted / LMF-based localization using a UE-side model, combined with AI / ML-assisted localization. In this case, location estimation is performed on the LMF side. However, the model is deployed on the UE side to execute an AI / ML-assisted localization method that estimates intermediate features. Here, the measurement is performed in the UE (downlink localization). This intermediate feature is reported to the LMF to calculate the location. A diagram of this case is shown in... Figure 4B The middle is represented.

[0137] Figure 4C Case 2b is illustrated—direct AI / ML localization using UE-assisted / LMF-based localization with an LMF-side model. The difference between this case and case 2a is that the model is deployed on the LMF side to enable the direct AI / ML localization method. As in the previous case, the measurement is performed in the UE (downlink localization). The illustration for this case is in... Figure 4C The middle is represented.

[0138] Figure 4D Case 3a is illustrated—NG-RAN node-assisted localization using a gNB-side model, combined with AI / ML-assisted localization. The difference between this case and case 2a is that the model is deployed on the gNB side to enable the AI / ML-assisted localization method. Another difference is that measurements are performed in the gNB (uplink localization). Here, the output is intermediate features, which should be reported to the LMF. Finally, the LMF performs location estimation. An illustration of this case is shown in... Figure 4D The middle is represented.

[0139] Figure 4E Case 3b is illustrated—NG-RAN node-assisted localization using the LMF-side model, directly AI / ML localization. In this case, the model output is the horizontal position, which is estimated in the LMF using measurements already obtained in the gNB (uplink localization) and the model deployed in the LMF, using direct AI / ML localization. The illustration of this case is in... Figure 4E The middle is represented.

[0140] In the following text, some terms used herein are defined based on the RAN1 protocol (source R1-2304148) regarding the list of terms used in AI / ML.

[0141] The following section lists the applicable conditions for locating use cases, based on Nokia proposal R1-2304680:

[0142] RAN1#112 has discussed the following FL proposals ([FL4] Proposals 5-8i) regarding applicable conditions: At least for the UE portion of both the UE-side and bilateral models, it is proposed to study: - How to define and study the set of applicable conditions for a function / [model], where applicable conditions can be used to enable the development / [site]-specific model for the scenario / configuration [and report the model’s suitability for the network when needed]. - Whether and how to define performance targets for the function / [model] (possibly as part of the applicable conditions), and - Whether and how the UE reports the applicable conditions (set) for the supported functions (and, if necessary, for the supported models) and / or the set of supported functions.

[0143] As stated in the FL proposal, RAN1 should first identify the applicable conditions for one or more functions supported for a given sub-use case (ML-enabled feature). In function identification and function-based LCM, prior to any other steps, the first step at the network level needs to understand the UE conditions (including parameters / configuration), as this will reveal the contextual conditions when using the ML model to support a given ML-enabled feature. These applicable conditions may vary depending on the different sub-use cases, but we at least expect a common set of applicable conditions (i.e., the definitions of parameters are common, but parameter values ​​may vary depending on the function or feature) to be derived across all sub-use cases discussed in Rel-18.

[0144] Figure 5 The features enabling ML are shown, particularly the use of functional and model IDs, as well as applicable condition information. That is, in Figure 5 The potential uses of the applicable conditions are shown in the diagram. Each function can be identified by a subset of the applicable conditions, but this does not prevent individually applicable conditions from being reused in different sets and / or different functions.

[0145] In the following text, the technical problems addressed by some example embodiments will be described.

[0146] To clarify the background of the technical problem to be solved, it should be noted that, from the perspective of ML positioning use cases, ML models can be deployed on the UE side, gNB side, or LMF side. Depending on the specific situation, DL PRS or UL SRS are used as reference signals for positioning measurements. Furthermore, "positioning measurements" or "intermediate features" are used (within the UE or network) to calculate the UE's final location.

[0147] When a UE moves within an NR network, it physically moves. This movement requires the UE to be served by different cells / TRPs. To support location services, the LMF needs to prepare (multiple) gNBs: - Configure radio resources for the UE, allowing the UE to measure and report the set of DL PRS for a given set of candidate cells or TRPs; and - Configure radio resources for the UE to transmit and report UL SRS measurements taken by a given set of candidate cells or TRPs.

[0148] LMF can prepare N such TRPs, which can span more than one gNB.

[0149] In actual deployment, especially for (the above reference) Figure 4DIn scenario 3a, for ML-based localization, the different (multiple) gNBs / (multiple) TRPs involved in localization determination can carry one or more ML models or (multiple) ML functions. Due to practical limitations, different (multiple) gNBs can carry different (multiple) ML models or (multiple) ML functions, resulting in different sets of "intermediate features." While this is a common operating pattern and ideal in actual deployments, it also introduces some practical problems to ML-based localization:

[0150] Question 1: Due to mismatches in ML functionality between (multiple) source cells and (multiple) neighboring cells, the concept of "intermediate features" may not be seamlessly integrated across (multiple) (TRPs) or (multiple) gNBs. This can impact ML positioning latency when extremely high accuracy and extremely low latency are required (e.g., 20 msec latency with cm-level accuracy for XR / VR services).

[0151] Question 2: Aspects of coordinating ML functions between (multiple) source cells and (multiple) neighboring cells in ML location use cases—for example, how to align the differences in (ML-configured) functions between the ML models of the source and target cells.

[0152] Some example implementations are intended to provide solutions to the above problems and define signaling procedures suitable for handling the above.

[0153] In the following sections, a general overview of some example embodiments is provided by reference. Figure 1A , 1B 2A and 2B are described.

[0154] Figure 1A An LMF 1 according to this example embodiment is shown. LMF 1 is an example for a device that can be part of a network control element, for example, performing location management functions. The processes performed by LMF 1 are... Figure 1B It is shown in the middle. Figure 1A The LMF 1 shown includes at least one processor 11 and at least one memory 12, the at least one memory 12 storing instructions that, when executed by at least one processor 11, cause the device to: determine a location configuration request based on the machine learning capabilities of a plurality of network nodes (e.g., gNB, TRP) capable of performing location-related measurements, the location configuration request indicating at least one location configuration, the at least one location configuration indicating at least one location method to be applied to estimate the location of the user equipment. Figure 1B In S11, the location configuration request is sent to at least one of the multiple network nodes. Figure 1BIn S12), at least one response is received from at least one of a plurality of network nodes, wherein the at least one response includes at least one selected location configuration ( Figure 1B S13 in the above), and based on the response received from the network node, configure the location configuration to multiple network nodes ( Figure 1B (S14 in the text).

[0155] Figure 2A Node 2 is shown according to this example embodiment. Node 2 is an example for a device and can be part of a network node, such as a serving node, a neighboring node, or an auxiliary node. For example, the node could be a gNB or a TRP. The process performed by node 2 is... Figure 2B It is shown in the middle. Figure 2A The LMF 2 shown includes at least one processor 21 and at least one memory 22, the at least one memory 22 storing instructions that, when executed by the at least one processor 21, cause the device to: receive a location configuration request from a network control element (e.g., LMF 1), the at least one location configuration request indicating at least one location configuration, the at least one location configuration including at least a location method to be applied to estimate the location of the user equipment. Figure 2B In S21), based on the device's machine learning capabilities, at least one location configuration of the location configuration request is selected. Figure 2B In S22), a response including at least one selected positioning configuration is sent to the network control element. Figure 2B In S23), information about the location configuration to be applied is received. Figure 2B (S24 in the original text), and based on the positioning configuration to be applied, performing positioning-related measurements using a machine learning model ( Figure 2B (S25 in the middle).

[0156] Therefore, according to the example embodiment, a network control element such as LMF 1 can align the ML functions applied to different network nodes for locating the UE. Thus, there is no mismatch in ML functions between different nodes.

[0157] like Figure 1A and 2A The devices 1 and 2 shown may include more components than described above, and may also include I / O units 13, 23 capable of sending to and receiving from other network elements.

[0158] LMF 1 can determine the ML-related capabilities of a network node among multiple network nodes by requesting ML-related capabilities from the network node and receiving a response from the network node including information about the ML-related capabilities. For example, this can be performed when a network node is added or updated.

[0159] Network nodes can be AI / ML enabled.

[0160] Network nodes can also indicate their traditional capabilities (e.g., the ability to combine intermediate features across ML-enabled and non-ML-enabled components).

[0161] LMF 1 can select network nodes that will participate in the location user equipment's ML-related capabilities based on the ML-related capabilities of multiple network nodes to which the location configuration request is sent.

[0162] LMF 1 can configure all network nodes in a network with the same positioning configuration, or configure individual positioning configurations for each network node. For example, LMF 1 can apply different positioning configurations based on the different ML or non-ML capabilities of network nodes, while ensuring that the results achieved by network nodes based on different positioning configurations can be combined to estimate the UE's location.

[0163] In the following text, some example embodiments are described in more detail.

[0164] For case 3a—NG-RAN-assisted AIML localization with intermediate features reported to the LMF (via LPP)—the following novel aspects are defined by some example embodiments: - Alignment of ML functions between (multiple) serving cells and (multiple) neighboring cells - Negotiate the format of the intermediate features generated by each gNB - Alignment of UL SRS based on the alignment of ML functions between (multiple) serving cells and (multiple) neighboring cells. - Combine "intermediate features" at LMF.

[0165] A more detailed signal flow is shown in Figure 6, which is composed of... Figure 6A and Figure 6B The components include: LMF, AIML-enabled serving nodes, (multiple) AIML-enabled neighboring cell nodes, and signal exchange between UEs. "AIML-enabled" means that the corresponding node applies AIML (or ML) to its operation; more specifically, it applies ML models or ML functions to perform location-related measurements, etc. In the following text, AIML-enabled serving nodes are also referred to as "serving nodes," and can be, for example, gNBs. AIML-enabled (multiple) neighboring nodes are also referred to as "(multiple) neighboring nodes," and can be, for example, TPRs.

[0166] In process A1, the LMF sends an NRRPa TRP information request, including an ML location information request, to the serving node. In process A2, the LMF sends a similar request to (multiple) neighboring nodes. In process A3, the serving node responds with an NRRPa TRP information response, which includes an ML location information response. In process A4, the LMF receives similar responses from (multiple) neighboring nodes.

[0167] Through the ML location information requests sent in processes A1 and A2, the LMF requests the ML-related capabilities (i.e., applicable conditions) of each gNB / TRP (i.e., from the serving node and (multiple) neighboring nodes), which include at least: • Supported intermediate features (for two-step localization), i.e., the input and output of its ML model: o LOS / NLOS indication o ToA, Path Phase o etc. • Supported UL SRS configurations (or UL SRS characteristics) for its measurements, i.e., inputs to its ML model: o UL SRS bandwidth, periodicity o Comb size, comb offset o etc. o CIR o Power Delay Distribution (PDP) o Delay distribution (DP) • ML model complexity (to account for any latency in localization) o Model size o Handling latency o etc.

[0168] In processes A3 and A4, the node (gNB / TRP) responds to the LMF by indicating the following: • Requested information (Process 1 and Process 2) • No ML-related capabilities o indicates its traditional capabilities (the combination of intermediate features enables it to cross ML-enabled and non-ML-enabled components). •...

[0169] Processes A3 and A4 may include additional combinations of capabilities supported by the node. An example is shown in the table below, which lists the combinations of capabilities published by the node to the LMF.

[0170] In procedure A5, the LMF updates the ML positioning configuration combination. That is, in procedure A5, the LMF determines the positioning method and UL SRS configuration based on the received information, taking into account the different ML capabilities at different candidate gNB / TRPs for positioning at the target UE. For example, according to the table above, the LMF can select capability combination N to configure the nodes for ML positioning features.

[0171] In process A6, the LMF performs LPP capability transmission, in which the LMF obtains UE capabilities from the UE.

[0172] In procedure A7, the LMF refines its determination by taking into account the UE capabilities (e.g., supported UL SRS bandwidth) obtained in procedure A6 via the regular LPP capability information exchange process. When a UE is restricted from having capabilities exceeding a given combination of capabilities, the LMF considers the specific UE's capabilities and adjusts the capability combinations in its database accordingly.

[0173] In procedures A8 and A9, the LMF provides the determined configuration to the nodes (serving node and (multiple) neighboring nodes) that want to participate in locating the target UE. In procedures A10 and A11, the LMF provides the nodes with a list of configuration options and collects the response.

[0174] In other words, in process A8, the LMF sends an NRPPa POS information request, including an MLPositioning Config Req, to ​​the serving node (gNB); and in process A9, the LMF sends a similar request to (multiple) serving nodes. In process A10, the LMF receives an NRPPa POS information response, including an MLPositioning Config RSP, from the serving node (gNB); and in process A11, the LMF receives a similar response from (multiple) serving nodes.

[0175] In one example implementation, LMF generates the following three configurations:

[0176] Configuration 1: LOS / NLOS indicator = YES, ToA = YES, SRS BW = 100 MHz, SRS period = 20 ms, comb size = 1, comb offset = 1, CNN with 16 layers

[0177] Configuration 2: LOS / NLOS indicator = YES, ToA = NO, SRS BW = 200 MHz, SRS period = 20 ms, comb size = 2, comb offset = 2, RNN with 16 layers

[0178] Configuration 3: LOS / NLOS indication = YES, ToA = NO, SRS BW = 400 MHz, SRS period = 20ms, comb size = 4, comb offset = 4, ResNet has 32 layers

[0179] It should be noted that a definition such as "ToA=YES" indicates that the reporting of ToA is supported, while a definition of "ToA=NO" indicates that it is not supported.

[0180] In processes A10 and A11, nodes can respond using preferred configuration options. For example, a service node can respond that it supports both configuration 1 and configuration 2, while a secondary node (i.e., one of the neighboring nodes) only supports configuration 2 and configuration 3. In this case, LMF will only configure configuration ID 2.

[0181] In process A12, gNB (service node) determines UL SRS resources.

[0182] In process A13, LMF updates the ML positioning configuration combination; and in process A14, LMF activates the given ML positioning configuration.

[0183] In procedure A15, the LMF sends an NRPPa POS activation request, including a configuration ID, to the serving node and (multiple) neighboring nodes. The configuration ID indicates the configuration selected by the LMF.

[0184] For example, in process 15, after combining the responses from processes 13 and 14, LMF will only configure configuration ID 2 (i.e., configuration 2 as specified above).

[0185] In procedure A16, the serving node instructs the UE to activate UE SRS transmission.

[0186] In procedure A17, the serving node sends an NRPPA POS activation response to the LMF. In procedure A18, the LMF sends an NRPPA measurement request to the serving node, and in procedure A19, the LMF sends NRPPA measurement requests to (multiple) neighboring nodes. Subsequently, the serving node and (multiple) neighboring nodes perform US SRS measurements in procedure 20, and in procedures A21 and A22, they send the results to the LMF as NRPPA measurement responses.

[0187] Therefore, in procedures A16 to A22, the UL SRS transmission is generated by the serving node and sent to the UE. The LMF requests the NRPPa measurement of the UL SRS from both the serving node and the auxiliary node.

[0188] In process A23, LMF combines intermediate features and derives the final position.

[0189] Because the localization configuration is aligned between different anchor nodes (e.g., including service nodes and neighboring nodes (e.g., gNBs or TRPSs) as described above), LMF is able to use an algorithm to combine intermediate features across nodes. This algorithm can simply combine samples across service and auxiliary nodes, or it can choose a weighting method, for example, considering that the measurements / features of the service gNB have a 50% weight, and the measurements / features of neighboring nodes have a 50% weight.

[0190] In mobile scenarios, as the UE moves, the serving gNB changes over time, and the LMF can adjust the weighted measurements or intermediate features collected from different gNBs accordingly, for example, by obtaining serving cell information from the AMF. Overall, due to the combination of intermediate features, the overall positioning features during the HO period can continue to function even when the UE moves away from the serving cell.

[0191] It should be noted that processes A1 through A23 need not be executed in the order described above. In particular, it should be noted that processes A1 through A5, which relate to obtaining ML-related capabilities from the service node and its neighboring nodes(s), can be executed separately. Once the LMF knows the ML-related capabilities of a node, it does not need to request these capabilities again. For example, these processes can be executed only when a new node is created or a node is updated.

[0192] Therefore, based on the above, and according to some example embodiments, the following is provided:

[0193] According to some embodiments, the Location Management Function (LMF) requests machine learning capabilities from (multiple) serving nodes and / or neighboring nodes. Upon receiving the capability response, the LMF determines a list of location configurations. The LMF then sends this list of location configurations to (multiple) serving nodes and / or neighboring nodes to locate the target UE, and receives responses from the nodes with selected / preferred configurations from the configuration list. Based on this, the LMF configures the location configuration to (multiple) serving nodes and / or neighboring nodes.

[0194] According to some embodiments, a node (which can be a serving node or a neighboring / auxiliary node) receives a machine learning capability request from the LMF and sends a response of (multiple) machine learning capabilities to the LMF. The node then receives a location configuration list and sends the selected / preferred location configuration from the location configuration list to the LMF. The node then receives the location configuration.

[0195] According to some embodiments, the LMF can request machine learning capabilities for each gNB / TRP. These machine learning capabilities may include: supported intermediate features (for two-step localization), i.e., the inputs and outputs of its ML model, such as LOS / NLOS indication, ToA, path phase, etc.; supported UL SRS configurations for its measurements, i.e., the inputs of its ML model, such as UL SRS bandwidth, periodicity, comb size, comb offset, etc.; and ML model complexity (to account for any latency in localization), such as model size, processing latency, etc.

[0196] If available, the capability response may include the requested information.

[0197] LMF can determine the positioning configuration for locating the target UE based on the machine learning capabilities of different candidate gNB / TRPs.

[0198] The positioning configuration can also take into account UE capabilities, such as the supported UL SRS bandwidth obtained via the traditional LPP capability information exchange process.

[0199] LMF can request NRPPa measurements of UL SRS from both the serving node and adjacent (auxiliary) nodes.

[0200] Service nodes and adjacent (auxiliary) nodes can be AI / ML enabled.

[0201] Additionally, capability response may include a combination of capabilities supported by the node.

[0202] The example embodiments described above are merely examples and may be modified.

[0203] For example, according to some of the example embodiments described above, LMF 1 can configure all network nodes in a plurality of network nodes with the same location configuration. However, the embodiments are not limited thereto. LMF 1 can also configure network nodes individually with individual location configurations. For example, although LMF 1 requests / configures the same output format for ML-based estimation at network nodes (gNB, TRP), it can, for example, configure (or request) different types of input for estimation configuration at different network nodes, depending on the input capabilities of the network nodes (e.g., measurement bandwidth).

[0204] Similarly, in this case, it is possible to ensure the coordination or at least combination of multiple network nodes supporting ML for a single estimation task (i.e., a single target UE).

[0205] The names of network elements, protocols, and methods are based on the current standard. In other versions or other technologies, the names of these network elements and / or protocols and / or methods may differ, as long as they provide the corresponding functionality.

[0206] Generally, the exemplary embodiments may be implemented by computer software stored in the memories (memory resources, memory circuitry) 12, 22 and executable by the processors (processing resources, processing circuitry) 11, 21, or by hardware, or by a combination of software and / or firmware and hardware.

[0207] The terms “connection,” “coupling,” or any variation thereof refer to any direct or indirect connection or coupling between two or more elements, and may include the presence of one or more intermediate elements between the two elements that are “connected” or “coupled.” The coupling or connection between elements can be physical, logical, or a combination thereof. As adopted herein, as a non-limiting example, two elements can be considered to be “connected” or “coupled” together by means of one or more wires, cables, and printed electrical connections, and by means of electromagnetic energy (such as electromagnetic energy having wavelengths in the radio frequency region, the microwave region, and the optical region (both visible and invisible light).

[0208] Memory (memory resources, memory circuitry) 12, 22 can be of any type suitable for the local technical environment and can be implemented using any suitable data storage technology, such as semiconductor-based memory devices, magnetic memory devices and systems, optical memory devices and systems, fixed and removable memory, and non-transitory computer-readable media. Processor (processing resources, processing circuitry) 11, 21 can be of any type suitable for the local technical environment and, by way of non-limiting example, can include one or more of a general-purpose computer, a special-purpose computer, a microprocessor, a digital signal processor (DSP), and a processor based on a multi-core processor architecture.

[0209] Furthermore, as used in this application, the term "circuit system" may refer to one or more of the following: (a) Hardware circuit implementation only (such as implementation only in analog and / or digital circuit systems) and (b) A combination of hardware circuitry and software, such as (if applicable): (i) A combination of (multiple) analog and / or digital hardware circuits and software / firmware, and (ii) Any part of a hardware processor (including (multiple) digital signal processors), software, and (multiple) memories, which work together to enable a device (such as a mobile phone or server) to perform various functions and (c) (Multiple) hardware circuits and / or (multiple) processors (such as (multiple) microprocessors or a portion of (multiple) microprocessors) that require software (e.g., firmware) to operate, but may not exist when operation does not require software.

[0210] This definition of "circuit system" applies to all uses of the term in this application, including in any claim. As another example, as used in this application, the term "circuit system" also covers only hardware circuitry or a processor (or multiple processors) or portions of hardware circuitry or processors and their accompanying software and / or firmware implementations. For example, if applicable to a particular claim element, the term "circuit system" also covers baseband integrated circuits or processor integrated circuits for mobile devices or similar integrated circuits in servers, cellular network devices, or other computing or network devices.

[0211] As used herein, the term “non-transient” refers to the limitations of the medium itself (i.e., tangible, not signaling), rather than limitations on the persistence of data storage (e.g., RAM vs. ROM).

[0212] It should be noted that, as used herein, “at least one of the following: ” and “at least one of ” and similar wording, where the list of two or more elements is connected by “and” or “or”, means at least any one element, or at least any two or more elements, or at least all elements.

[0213] It should be understood that the various exemplary embodiments of this disclosure are illustrative only and not restrictive, and should not be construed as limiting. Various modifications and applications will be apparent to those skilled in the art without departing from the spirit and scope of the various exemplary embodiments of this disclosure.

Claims

1. An apparatus comprising: At least one processor, and At least one memory storing instructions that, when executed by the at least one processor, cause the device to at least: Based on the machine learning capabilities of multiple network nodes capable of performing location-related measurements, a location configuration request is determined, the location configuration request indicating at least one location configuration, the at least one location configuration including at least a location method to be applied to estimate the location of the user equipment; Send the location configuration request to at least one of the plurality of network nodes; Receive at least one response from at least one of the plurality of network nodes, wherein the at least one response includes at least one selected location configuration; as well as Based on the at least one response received from at least one of the plurality of network nodes, at least one location configuration is configured to at least one of the plurality of network nodes.

2. The apparatus of claim 1, wherein the instructions stored in the at least one memory, when executed by the at least one processor, further cause the apparatus to at least: The machine learning capabilities of the network nodes among the plurality of network nodes are determined by the following: Request information from the network node about its machine learning capabilities, and Receive a response from the network node, the response including the information regarding the machine learning-related capabilities.

3. The apparatus of claim 1 or 2, wherein the machine learning-related capabilities include information about the machine learning models and / or machine learning functions applied by each of the plurality of network nodes.

4. The apparatus of claim 3, wherein the information relating to the machine learning model and / or machine learning function includes: The type, characteristics, features and / or structure of the machine learning model and / or machine learning function, and / or The inputs and / or outputs of the machine learning model and / or machine learning function, and / or Information regarding the computational complexity and / or interference latency of the machine learning model and / or machine learning function.

5. The apparatus according to claim 4, wherein The output of the machine learning model and / or machine learning function includes at least one supported intermediate feature.

6. The apparatus according to any one of claims 1 to 5, wherein the at least one positioning configuration indicated by the positioning configuration request further comprises at least one of the following: The machine learning model and / or machine learning function to be applied; The inputs and / or outputs of the machine learning model and / or machine learning function; and Uplink detection reference signal configuration.

7. The apparatus of claim 6, wherein the instructions stored in the at least one memory, when executed by the at least one processor, further cause the apparatus to at least: Based on the machine learning capabilities of the multiple network nodes, the location configuration request to the multiple network nodes is determined as follows: Alignment will be achieved by machine learning models and / or machine learning functions applied by the multiple network nodes; Alignment will be applied to the format of the intermediate features of the aligned machine learning model; as well as Align uplink probe reference signal configuration.

8. The apparatus according to any one of claims 1 to 7, wherein the instructions stored in the at least one memory, when executed by the at least one processor, further cause the apparatus to at least: Based on the responses received from all of the plurality of network nodes, the positioning configuration is configured for at least one of the plurality of network nodes as follows: Determine which location configuration or which location configurations are supported by all of the plurality of network nodes.

9. The apparatus according to any one of claims 1 to 8, wherein the instructions stored in the at least one memory, when executed by the at least one processor, further cause the apparatus to at least: Configure all network nodes of the plurality of network nodes to have the same positioning configuration, or Each of the plurality of network nodes is individually configured with a location configuration.

10. The apparatus according to any one of claims 1 to 9, wherein the instructions stored in the at least one memory, when executed by the at least one processor, further cause the apparatus to at least: Receive positioning-related measurement results from the multiple network nodes, and The location of the user equipment is determined based on the combination of the received positioning-related measurement results.

11. The apparatus of claim 10, wherein the instructions stored in the at least one memory, when executed by the at least one processor, further cause the apparatus to at least: The intermediate features are combined by applying weights to the location-related measurement results.

12. The apparatus of claim 11, wherein the instructions stored in the at least one memory, when executed by the at least one processor, further cause the apparatus to at least: The weights of the intermediate features received from the plurality of network nodes are adjusted based on the movement of the user equipment.

13. The apparatus according to any one of claims 10 to 12, wherein the positioning-related measurement results include intermediate features.

14. An apparatus comprising: At least one processor, and At least one memory storing instructions that, when executed by the at least one processor, cause the device to at least: A location configuration request is received from a network control element, the location configuration request indicating at least one location configuration, the at least one location configuration including at least a location method to be applied to estimate the location of a user equipment; Based on the machine learning capabilities of the device, at least one location configuration of the location configuration request is selected; Send a response to the network control element including at least one selected positioning configuration; Receive information about the location configuration to be applied; as well as Based on the location configuration to be applied, location-related measurements are performed using a machine learning model.

15. The apparatus of claim 14, wherein the instructions stored in the at least one memory, when executed by the at least one processor, further cause the apparatus to at least: Receive from the network control element a request for information regarding the machine learning capabilities of the device; and Send a response, including the information about machine learning-related capabilities, to the network control element.

16. The apparatus of claim 14 or 15, wherein the machine learning-related capabilities include: Information regarding the machine learning models and / or machine learning functions applied by the device.

17. The apparatus of claim 16, wherein the information relating to the machine learning model and / or machine learning function includes: The type, characteristics, features and / or structure of the machine learning model and / or machine learning function, and / or The inputs and / or outputs of the machine learning model, and / or Information regarding the computational complexity of the machine learning model and / or the interference latency of the machine learning model and / or machine learning function.

18. The apparatus of claim 17, wherein the output of the machine learning model and / or machine learning function includes at least one supported intermediate feature.

19. The apparatus according to any one of claims 14 to 18, wherein the at least one positioning configuration indicated by the positioning configuration request further comprises at least one of the following: The machine learning model and / or machine learning function to be applied; The inputs and / or outputs of the machine learning model and / or machine learning function; and Uplink detection reference signal configuration.

20. The apparatus according to any one of claims 14 to 19, wherein the instructions stored in the at least one memory, when executed by the at least one processor, further cause the apparatus to at least: The results of the positioning-related measurements are sent to the network control element.

Images