Machine learning for positioning

Abstract

Aspects of the present disclosure relate to methods, apparatuses, and systems for supporting machine learning for positioning. For example, implementations provide for artificial intelligence (AI)-based direct positioning and AI-assisted positioning, which may be utilized to improve location accuracy performance of a user equipment (UE). In example implementations, for direct AI / machine learning (ML) positioning, the AI / ML model may utilize techniques such as fingerprint recognition to achieve enhanced positional accuracy, such as via measurements and environmental data. Thus, the present disclosure provides techniques to configure direct AI / ML positioning assistance data and define measurements to perform AI / ML direct positioning. In addition, the present disclosure provides techniques to configure reporting criteria for nodes and / or other entities performing AI / ML direct positioning measurements.

Description

[0001] Related applications

[0002] This application claims priority to U.S. Provisional Application Serial No. 63 / 484,102, filed on February 9, 2023, entitled “MACHINE LEARNING FOR POSITIONING,” and U.S. Provisional Application Serial No. 63 / 444,469, filed on February 9, 2023, entitled “MACHINE LEARNING FOR POSITIONING,” the entire contents of which are incorporated herein by reference. Technical Field

[0003] The present disclosure relates to wireless communications, and more particularly to position determination in wireless communications. Background Art

[0004] A wireless communication system may include one or more network communication devices (such as base stations), which may also be referred to as eNodeBs (eNBs), next generation NodeBs (gNBs), or other appropriate terms. Each network communication device (such as a base station) may support wireless communication for one or more user communication devices, which may also be referred to as user equipment (UEs) or other appropriate terms. A wireless communication system may support wireless communication with one or more user communication devices by utilizing resources of the wireless communication system (e.g., time resources (e.g., symbols, time slots, subframes, frames, etc.) or frequency resources (e.g., subcarriers, carriers)). Additionally, a wireless communication system may support wireless communication across various radio access technologies, including third generation (3G) radio access technology, fourth generation (4G) radio access technology, fifth generation (5G) radio access technology, and other appropriate radio access technologies beyond 5G (e.g., sixth generation (6G)).

[0005] Some wireless communication systems provide methods for determining the location of a device (eg, a UE), such as the geographic location of the UE. However, current implementations for UE positioning may be inaccurate. Summary of the Invention

[0006] The present disclosure relates to methods, apparatus, and systems that support machine learning for positioning. For example, implementations provide direct positioning and AI / ML-assisted positioning based on artificial intelligence / machine learning (AI / ML), which can be utilized to improve UE location accuracy performance. In an example implementation, for direct AI / ML positioning, the AI / ML model can utilize techniques such as fingerprinting to obtain enhanced location accuracy, such as via measurements and environmental data. Accordingly, the present disclosure provides techniques for configuring direct AI / ML positioning assistance data, and defining measurements to perform AI / ML direct positioning. Furthermore, the present disclosure provides techniques for configuring reporting criteria for nodes and / or other entities that perform AI / ML direct positioning measurements.

[0007] Thus, by utilizing the described techniques, a more accurate positioning of a UE may be obtained, and usage of system resources used to determine the UE's position may be reduced.

[0008] Some implementations of the methods and apparatus described herein may also include sending one or more machine learning positioning report requests comprising a machine learning reporting configuration and one or more common reporting criteria; receiving one or more machine learning positioning reports; processing the one or more machine learning positioning reports via a machine learning model; and generating an estimated location of a user equipment (UE) based at least in part on an output from the machine learning model.

[0009] Some implementations of the methods and apparatus described herein may further include: wherein the machine learning report configuration includes one or more of the following items: direct machine learning report configuration, or assisted machine learning report configuration; wherein the method is performed by an apparatus, the apparatus including a configuration entity, and wherein the configuration entity includes at least one of the following items: a location server, a next generation radio access network (NG-RAN), a positioning reference unit (PRU), a UE, an anchor UE, or a target UE; the method further includes: sending one or more machine learning positioning report requests from the first apparatus to one or more second apparatuses, and receiving one or more machine learning positioning reports from the one or more second apparatuses, and wherein the one or more second apparatuses include at least one of the following items: a location server, a next generation radio access network (NG-RAN), a positioning reference unit (PRU), a UE, an anchor UE, or a target UE.

[0010] Some implementations of the methods and apparatus described herein may further include: wherein one or more public reporting standards include one or more of the following: fingerprint type, ground truth reference location type, time domain reporting type, measurement preprocessing, measurement validity, UE type, mobility indication, orientation indication, fingerprint recognition environment indication, fingerprint quality indication, or tag quality indication; wherein for at least one machine learning positioning report request, one or more of the public reporting standards are configured to be enhanced, removed, updated, activated, or deactivated; the method further includes: broadcasting one or more public reporting standards via positioning system information broadcast signaling; wherein the machine learning report configuration includes: an indication for reporting machine learning positioning measurement correlations between different measurement sets.

[0011] Some implementations of the methods and apparatus described herein may further include: wherein the machine learning positioning measurement correlation comprises one or more of: spatial correlation, or temporal correlation; wherein one or more machine learning positioning reports comprise one or more of: machine learning positioning measurements, or machine learning positioning location information; wherein the machine learning report configuration comprises: an indication for reporting path loss at different locations, the different locations comprising a ground truth reference location; wherein the machine learning report configuration comprises: an indication for averaging the machine learning positioning measurements over a configured number of measurements and reporting the average as part of one or more machine learning positioning reports; wherein the machine learning report configuration comprises: an indication for determining a similarity score at configured locations to determine an optimal mapping between the fingerprint measurements and the estimated location of the target UE.

[0012] Some implementations of the methods and apparatus described herein may further include receiving one or more machine learning positioning report requests comprising a machine learning reporting configuration and one or more public reporting standards; generating one or more machine learning positioning reports based at least in part on the machine learning reporting configuration and the one or more public reporting standards; and sending the one or more machine learning positioning reports.

[0013] Some implementations of the methods and apparatus described herein may also include: wherein the method is performed by an apparatus comprising one or more of the following: a user equipment (UE), an anchor UE, or a target UE; wherein the method is performed by an apparatus comprising a configuration entity, and wherein the configuration entity comprises at least one of the following: a location server, a next generation radio access network (NG-RAN), or a positioning reference unit (PRU).

[0014] Some implementations of the methods and apparatus described herein may also include sending one or more machine learning positioning report requests, the machine learning positioning report requests including a machine learning report configuration and one or more public reporting standards; receiving one or more machine learning positioning reports; generating a machine learning positioning training dataset based at least in part on the one or more machine learning positioning reports; and training a positioning machine learning model using the machine learning positioning training dataset to generate a trained positioning machine learning model.

[0015] Some implementations of the methods and apparatus described herein may further include: receiving one or more additional machine learning positioning reports; inputting at least a portion of the one or more additional machine learning positioning reports into a trained positioning machine learning model; and estimating a location of a target user equipment (UE) based at least in part on an output from the trained positioning machine learning model; the method further includes: receiving a request for machine learning positioning training data for positioning; and sending a machine learning positioning training data set based at least in part on the request; the method further includes: receiving a request for a trained positioning machine learning model; and sending at least a portion of the trained positioning machine learning model based at least in part on the request. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 Illustrated is an example of a wireless communication system supporting machine learning for positioning according to aspects of the present disclosure.

[0017] Figure 2 Illustrated is a system in which positioning reference signals may be utilized to obtain positioning measurements.

[0018] Figure 3 The diagram illustrates a scenario for multi-cell RTT positioning.

[0019] Figure 4a and Figure 4b Portions of the LPP RequestLocationInformation message are illustrated.

[0020] Figure 5a and Figure 5b The diagram shows parts of the LPP ProvideLocationInformation message.

[0021] Figure 6 Illustrated is a system for machine learning-based RAN intelligence.

[0022] Figure 7 Illustrated are example scenarios supporting machine learning for positioning according to aspects of the present disclosure.

[0023] Figure 8Illustrated are example scenarios supporting machine learning for positioning according to aspects of the present disclosure.

[0024] Figure 9 Illustrated are messages that support machine learning for positioning according to aspects of the present disclosure.

[0025] Figure 10a and Figure 10b Illustrated are different portions of a message that support machine learning for positioning according to aspects of the present disclosure.

[0026] Figure 11 Illustrated are messages that support machine learning for positioning according to aspects of the present disclosure.

[0027] Figure 12 Illustrated are messages that support machine learning for positioning according to aspects of the present disclosure.

[0028] Figure 13 Illustrated is a scenario supporting machine learning for positioning according to aspects of the present disclosure.

[0029] Figure 14a and Figure 14b Illustrated is a scenario supporting machine learning for positioning according to aspects of the present disclosure.

[0030] Figure 15 Illustrated is a scenario supporting machine learning for positioning according to aspects of the present disclosure.

[0031] Figure 16 Illustrated is a scenario supporting machine learning for positioning according to aspects of the present disclosure.

[0032] Figure 17 Illustrated is a scenario supporting machine learning for positioning according to aspects of the present disclosure.

[0033] Figure 18 and Figure 19 Illustrated is an example block diagram of a device supporting machine learning for positioning according to aspects of the present disclosure.

[0034] Figures 20 to 25 A flow chart of a method supporting machine learning for positioning according to aspects of the present disclosure is illustrated. DETAILED DESCRIPTION

[0035] In wireless communication systems, techniques are used to estimate the location of a UE (e.g., positioning), such as the UE's geographic location and / or the UE's relative network location. For example, some systems utilize beam-based attempts to estimate UE location, such as in commercial and regulatory (e.g., emergency) scenarios. However, current location determination techniques may be imprecise and result in an inaccurate indication of the UE's location.

[0036] Thus, the present disclosure provides techniques for supporting machine learning for positioning. For example, implementations provide AI-based direct positioning and AI-assisted positioning that can be utilized to improve UE location accuracy performance, such as within the positioning framework defined by 3GPP. For example, for direct AI / ML positioning, the AI / ML model can utilize techniques such as fingerprinting to obtain enhanced location accuracy, such as via measurements and environmental data. Thus, the present disclosure provides techniques for configuring direct AI / ML positioning assistance data, and defining measurements to perform AI / ML direct positioning. Furthermore, the present disclosure provides techniques for configuring reporting criteria for nodes and / or other entities performing AI / ML direct positioning measurements.

[0037] Therefore, by utilizing the described techniques, a more accurate positioning of the UE can be obtained by utilizing a large amount of radio and other related data, and the use of system resources for determining the UE position can be reduced.

[0038] Aspects of the present disclosure are described in the context of a wireless communication system.Aspects of the present disclosure are further illustrated and described with reference to device diagrams and flow charts.

[0039] Figure 1 An example of a wireless communication system 100 that supports machine learning for positioning according to aspects of the present disclosure is illustrated. The wireless communication system 100 may include one or more network entities 102, one or more UEs 104, a core network 106, and a packet data network 108. The wireless communication system 100 may support various radio access technologies. In some implementations, the wireless communication system 100 may be a 4G network, such as an LTE network or an Advanced LTE (LTE-A) network. In some other implementations, the wireless communication system 100 may be a 5G network, such as an NR network. In other implementations, the wireless communication system 100 may be a combination of a 4G network and a 5G network, or other suitable radio access technologies, including Institute of Electrical and Electronics Engineers (IEEE) 802.11 (Wi-Fi), IEEE 802.16 (WiMAX), IEEE 802.20. The wireless communication system 100 may support radio access technologies other than 5G. Additionally, the wireless communication system 100 may support technologies such as time division multiple access (TDMA), frequency division multiple access (FDMA), or code division multiple access (CDMA).

[0040] One or more network entities 102 may be dispersed throughout a geographic area to form a wireless communication system 100. One or more of the network entities 102 described herein may be, include, or be referred to as a network node, a base station, a network element, a radio access network (RAN), a base transceiver station, an access point, a NodeB, an eNodeB (eNB), a next generation NodeB (gNB), or other suitable terminology. The network entity 102 and the UE 104 may communicate via a communication link 110, which may be a wireless or wired connection. For example, the network entity 102 and the UE 104 may perform wireless communication (e.g., receive signaling, send signaling) over a Uu interface.

[0041] The network entity 102 may provide a geographic coverage area 112 for which it may support services (e.g., voice, video, packet data, messaging, broadcast, etc.) for one or more UEs 104 within the geographic coverage area 112. For example, the network entity 102 and the UEs 104 may support wireless communication of signals associated with the services (e.g., voice, video, packet data, messaging, broadcast, etc.) based on one or more radio access technologies. In some implementations, the network entity 102 may be mobile, such as a satellite associated with a non-terrestrial network. In some implementations, different geographic coverage areas 112 associated with the same or different radio access technologies may overlap, but different geographic coverage areas 112 may be associated with different network entities 102. The information and signals described herein may be represented using any of a variety of different technologies and techniques. For example, data, instructions, commands, information, signals, bits, symbols, and chips referenced throughout the description may be represented by voltages, currents, electromagnetic waves, magnetic fields or particles, optical fields or particles, or any combination thereof.

[0042] One or more UEs 104 may be dispersed throughout the geographic area of ​​the wireless communication system 100. The UE 104 may include or may be referred to as a mobile device, a wireless device, a remote device, a remote unit, a handheld device, a subscriber device, or some other suitable terminology. In some implementations, the UE 104 may be referred to as a unit, a station, a terminal, or a client, etc. Additionally or alternatively, the UE 104 may be referred to as an Internet of Things (IoT) device, an Internet of Everything (IoE) device, or a Machine Type Communication (MTC) device, etc. In some implementations, the UE 104 may be stationary within the wireless communication system 100. In some other implementations, the UE 104 may be mobile within the wireless communication system 100.

[0043] One or more UEs 104 may be devices of different forms or with different capabilities. Figure 1Some examples of UE 104 are shown in FIG. Figure 1 As shown, the UE 104 may be capable of communicating with various types of devices, such as a network entity 102, other UEs 104, or a network device (e.g., a core network 106, a packet data network 108, a relay device, an integrated access and backhaul (IAB) node, or another network device). Additionally or alternatively, the UE 104 may support communication with other network entities 102 or UEs 104 that may act as relays in the wireless communication system 100.

[0044] The UE 104 may also be capable of supporting wireless communications directly with other UEs 104 via a communication link 114. For example, the UE 104 may support wireless communications directly with another UE 104 via a device-to-device (D2D) communication link. In some implementations, such as vehicle-to-vehicle (V2V) deployments, V2X deployments, or cellular V2X deployments, the communication link 114 may be referred to as a sidelink. For example, the UE 104 may support wireless communications directly with another UE 104 via a PC5 interface.

[0045] The network entities 102 may support communication with the core network 106 or with another network entity 102, or both. For example, the network entities 102 may interface with the core network 106 via one or more backhaul links 116 (e.g., via S1, N2, N2, or another network interface). The network entities 102 may communicate with each other via the backhaul links 116 (e.g., via X2, Xn, or another network interface). In some implementations, the network entities 102 may communicate directly with each other (e.g., between the network entities 102). In some other implementations, the network entities 102 may communicate with each other or indirectly (e.g., via the core network 106). In some implementations, one or more network entities 102 may include subcomponents, such as an access network entity, which may be an example of an access node controller (ANC). The ANC may communicate with one or more UEs 104 via one or more other access network transport entities, which may be referred to as radio heads, smart radio heads, or transmission reception points (TRPs)).

[0046] In some implementations, the network entity 102 can be configured with a decomposed architecture that can be configured to utilize a protocol stack that is physically or logically distributed between two or more network entities 102, such as an integrated access backhaul (IAB) network, an open RAN (O-RAN) (e.g., a network configuration sponsored by the O-RAN Alliance), or a virtualized RAN (vRAN) (e.g., a cloud RAN (C-RAN)). For example, the network entity 102 can include one or more of the following: a central unit (CU), a distributed unit (DU), a radio unit (RU), a RAN intelligent controller (RIC) (e.g., a near real-time RIC (near real-time (RT) RIC), a non-real-time RIC (non-RT RIC)), a service management and orchestration (SMO) system, or any combination thereof.

[0047] The RU may also be referred to as a radio head, smart radio head, remote radio head (RRH), remote radio unit (RRU), or transmission reception point (TRP). In a disaggregated RAN architecture, one or more components of the network entity 102 may be collocated, or one or more components of the network entity 102 may be located in distributed locations (e.g., separate physical locations). In some implementations, one or more network entities 102 of the disaggregated RAN architecture may be implemented as virtual units (e.g., virtual CU (VCU), virtual DU (VDU), virtual RU (VRU)).

[0048] The functional split between CU, DU and RU can be flexible, and different functions can be supported depending on the functions performed at the CU, DU or RU (e.g., network layer functions, protocol layer functions, baseband functions, radio frequency functions and any combination thereof). For example, a functional split of the protocol stack can be adopted between the CU and the DU, so that the CU can support one or more layers of the protocol stack and the DU can support one or more different layers of the protocol stack. In some implementations, the CU can host upper protocol layer (e.g., Layer 3 (L3), Layer 2 (L2)) functions and signaling (e.g., Radio Resource Control (RRC), Service Data Adaptation Protocol (SDAP), Packet Data Convergence Protocol (PDCP)). The CU can be connected to one or more DUs or RUs, and one or more DUs or RUs can host lower protocol layers, such as Layer 1 (L1) (e.g., physical (PHY) layer) or L2 (e.g., radio link control (RLC) layer, media access control (MAC) layer) functions and signaling, and each DU or RU can be at least partially controlled by the CU.

[0049] Additionally or alternatively, a functional split of the protocol stack may be employed between the DU and the RU such that the DU may support one or more layers of the protocol stack and the RU may support one or more different layers of the protocol stack. The DU may support one or more different cells (e.g., via one or more RUs). In some implementations, the functional split between the CU and the DU or between the DU and the RU may be within the protocol layer (e.g., some functions for a protocol layer may be performed by one of the CU, DU, or RU, while other functions of the protocol layer may be performed by different items of the CU, DU, or RU).

[0050] The CU can be further functionally split into CU control plane (CU-CP) and CU user plane (CU-UP) functions. The CU can be connected to one or more DUs via a medium-range communication link (e.g., F1, F1-c, F1-u), and the DU can be connected to one or more RUs via a fronthaul communication link (e.g., an open fronthaul (FH) interface). In some implementations, the medium-range communication link or the fronthaul communication link can be implemented according to an interface (e.g., a channel) between layers of a protocol stack supported by the corresponding network entity 102 communicating via such a communication link.

[0051] The core network 106 may support user authentication, access authorization, tracking, connectivity, and other access, routing, or mobility functions. The core network 106 may be an evolved packet core (EPC) or a 5G core (5GC), which may include control plane entities that manage access and mobility (e.g., mobility management entity (MME), access and mobility management function (AMF)), and user plane entities that route packets or interconnections to external networks (e.g., serving gateway (S-GW), location management function (LMF) (which is a control plane entity that manages location-related services), packet data network (PDN) gateway (P-GW), or user plane function (UPF)). In some implementations, the control plane entity may manage non-access stratum (NAS) functions, such as mobility, authentication, and bearer management (e.g., data bearer, signaling bearer, etc.) for one or more UEs 104 served by one or more network entities 102 associated with the core network 106.

[0052] The core network 106 can communicate with the packet data network 108 via one or more backhaul links 116 (e.g., via S1, N2, N2, or another network interface). The packet data network 108 can include an application server 118. In some implementations, one or more UEs 104 can communicate with the application server 118. The UE 104 can establish a session (e.g., a PDU session, etc.) with the core network 106 via the network entity 102. The core network 106 can use the established session (e.g., the established PDU session) to route traffic (e.g., control information, data, etc.) between the UE 104 and the application server 118. The PDU session can be an example of a logical connection between the UE 104 and the core network 106 (e.g., one or more network functions of the core network 106).

[0053] In the wireless communication system 100, the network entity 102 and the UE 104 may use the resources of the wireless communication system 100 (e.g., time resources (e.g., symbols, time slots, subframes, frames, etc.) or frequency resources (e.g., subcarriers, carriers)) to perform various operations (e.g., wireless communications). In some implementations, the network entity 102 and the UE 104 may support different resource structures. For example, the network entity 102 and the UE 104 may support different frame structures. In some implementations, such as in 4G, the network entity 102 and the UE 104 may support a single frame structure. In some other implementations, such as in 5G and other suitable radio access technologies, the network entity 102 and the UE 104 may support various frame structures (e.g., multiple frame structures). The network entity 102 and the UE 104 may support various frame structures based on one or more digital technologies.

[0054] One or more digital technologies may be supported in the wireless communication system 100, and the digital technologies may include subcarrier spacing and cyclic prefixes. A first digital technology (e.g., μ = 0) may be associated with a first subcarrier spacing (e.g., 15 kHz) and a normal cyclic prefix. A first digital technology (e.g., μ = 0) associated with a first subcarrier spacing (e.g., 15 kHz) may utilize one slot per subframe. A second digital technology (e.g., μ = 1) may be associated with a second subcarrier spacing (e.g., 30 kHz) and a normal cyclic prefix. A third digital technology (e.g., μ = 2) may be associated with a third subcarrier spacing (e.g., 60 kHz) and a normal cyclic prefix or an extended cyclic prefix. A fourth digital technology (e.g., μ = 3) may be associated with a fourth subcarrier spacing (e.g., 120 kHz) and a normal cyclic prefix. A fifth digital technology (e.g., μ = 4) may be associated with a fifth subcarrier spacing (e.g., 240 kHz) and a normal cyclic prefix.

[0055] The time intervals of resources (e.g., communication resources) can be organized according to frames (also referred to as radio frames). Each frame can have a duration, for example, a duration of 10 milliseconds (ms). In some implementations, each frame can include multiple subframes. For example, each frame can include 10 subframes, and each subframe can have a duration, for example, a duration of 1 ms. In some implementations, each frame can have the same duration. In some implementations, each subframe of a frame can have the same duration.

[0056] Additionally or alternatively, the time intervals of resources (e.g., communication resources) can be organized according to time slots. For example, a subframe may include a certain number (e.g., quantity) of time slots. Each time slot may include a certain number (e.g., quantity) of symbols (e.g., orthogonal frequency division multiplexing (OFDM) symbols). In some implementations, the number (e.g., quantity) of time slots for a subframe may depend on the digital technology. For a conventional cyclic prefix, a time slot may include 14 symbols. For an extended cyclic prefix (e.g., for a 60kHz subcarrier spacing), a time slot may include 12 symbols. For a conventional cyclic prefix and an extended cyclic prefix, the relationship between the number of symbols per time slot, the number of time slots per subframe, and the number of time slots per frame may depend on the digital technology. It should be understood that references to a first digital technology (e.g., μ=0) associated with a first subcarrier spacing (e.g., 15kHz) may be used interchangeably between subframes and time slots.

[0057] In the wireless communication system 100, the electromagnetic (EM) spectrum can be divided into various categories, frequency bands, frequency channels, etc. based on frequency or wavelength. For example, the wireless communication system 100 can support one or more operating frequency bands, such as the frequency range designations FR1 (410 MHz-7.125 GHz), FR2 (24.25 GHz-52.6 GHz), FR3 (7.125 GHz-24.25 GHz), FR4 (52.6 GHz-114.25 GHz), FR4a or FR4-1 (52.6 GHz-71 GHz), and FR5 (114.25 GHz-300 GHz). In some implementations, the network entity 102 and the UE 104 can perform wireless communications on one or more operating frequency bands. In some implementations, FR1 can be used by the network entity 102 and the UE 104, as well as other devices or apparatuses, for cellular communication traffic (e.g., control information, data). In some implementations, FR2 can be used by the network entity 102 and the UE 104, as well as other devices or apparatuses, for short-range, high data rate capabilities.

[0058] FR1 may be associated with one or more digital technologies (e.g., at least three digital technologies). For example, FR1 may be associated with a first digital technology (e.g., μ = 0) including a subcarrier spacing of 15 kHz, a second digital technology (e.g., μ = 1) including a subcarrier spacing of 30 kHz, and a third digital technology (e.g., μ = 2) including a subcarrier spacing of 60 kHz. FR2 may be associated with one or more digital technologies (e.g., at least two digital technologies). For example, FR2 may be associated with a third digital technology (e.g., μ = 2) including a subcarrier spacing of 60 kHz, and a fourth digital technology (e.g., μ = 3) including a subcarrier spacing of 120 kHz.

[0059] In accordance with an implementation of machine learning for positioning, the network entity 102 sends an ML positioning configuration to the UE 104. For example, the ML positioning configuration 120 includes various information and / or parameters for the UE 104 to measure various positioning information, generate positioning measurements, and / or process the positioning measurements. Thus, based at least in part on the ML positioning configuration 120, the UE 104 performs ML positioning measurements 122, such as to measure and process various properties of wireless signals detected at the UE 104. The UE 104 generates an ML positioning response 124 based at least in part on the ML positioning measurements 122 and / or the ML positioning configuration 120, and sends the ML positioning response to the network entity 102. In an implementation, the ML positioning response 124 may include an estimated position (e.g., a location) of the UE 104, and / or the network entity 102 may process the ML positioning response using ML techniques to estimate the position of the UE 104.

[0060] In some wireless communication systems, NR positioning based on NR Uu signals and standalone (SA) architecture (e.g., beam-based transmission) are specified. Target use cases include commercial and regulatory (emergency services) scenarios. Performance parameters include the following [Technical Report (TR) 38.855]:

[0061] Positioning error Indoor Outdoor Horizontal positioning For 80% of UEs < 3m For 80% of UEs < 10m Vertical positioning For 80% of UEs < 3m For 80% of UEs < 3m

[0062] Additionally, some systems specify positioning performance parameters for commercial and IIoT use cases as follows [TR 38.857]:

[0063]

[0064] At least some of the supported positioning technologies are shown in Table 1 [TS 38.305]:

[0065] Table 1

[0066]

[0067]

[0068] Currently, the individual positioning techniques shown in Table 1 can be configured and performed based on the requirements of the LMF and UE capabilities. The transmission of the Positioning Reference Signal (PRS) enables the UE to perform UE positioning related measurements to be able to calculate the UE's position estimate and is configured per Transmit Reception Point (TRP), where a TRP can transmit one or more beams.

[0069] Figure 2 Illustrated is a system 200 in which positioning reference signals can be utilized to obtain positioning measurements. For example, PRS can be transmitted by different base stations (serving base station and neighboring base station) using narrow beams in FR1 and FR2, which is relatively different from LTE where PRS is transmitted throughout the cell. PRS can be locally associated with a PRS resource identifier (ID) and resource set ID for the base station (TRP). Similarly, UE positioning measurements (such as reference signal time difference (RSTD) and PRS RSRP measurements) are made between beams (e.g., between different downlink (DL) PRS resource pairs or DL ​​PRS resource sets) rather than between different cells as in LTE. In addition, the network utilizes additional UL positioning methods to calculate the position of the target UE.

[0070] Tables 2 and 3 show the reference signal to measurement mapping required for each of the RAT-dependent positioning techniques supported at the UE and gNB, respectively. RAT-dependent positioning techniques involve 3GPP RAT and core network entities to perform UE position estimation, in contrast to RAT-independent positioning techniques that rely on GNSS, inertial measurement unit (IMU) sensors, WLAN, and Bluetooth technologies for performing target device (UE) positioning.

[0071] Table 2: UE measurements used to implement RAT-related positioning techniques

[0072]

[0073]

[0074] Table 3: gNB measurements used to implement RAT-related positioning techniques

[0075]

[0076] The following RAT-related positioning technologies [TS38.305] can be supported:

[0077] The downlink time difference of arrival (DL-TDOA) positioning method utilizes the DL reference signal time difference (RSTD) (and optionally the DL PRS reference signal received power (RSRP)) of downlink signals received at the UE from multiple TPs. The UE uses assistance data received from a positioning server to measure the DL RSTD (and optionally the DL PRS RSRP) of the received signal, and the resulting measurements, along with other configuration information, are used to position the UE relative to neighboring TPs.

[0078] The DL AoD positioning method utilizes the measured DL PRS RSRP of downlink signals received at the UE from multiple TPs. The UE uses assistance data received from a positioning server to measure the DL PRS RSRP of the received signals, and the resulting measurements are used together with other configuration information to position the UE relative to neighboring TPs.

[0079] Figure 3 A scenario 300 for multi-cell round trip time (RTT) positioning is illustrated. The multi-RTT positioning method utilizes UE Rx-Tx measurements and DL PRS RSRP of downlink signals received from multiple TRPs, as measured by the UE, and gNB Rx-Tx measurements and uplink (UL) sounding reference signal (SRS)-RSRP of uplink signals transmitted from the UE measured at multiple TRPs. The UE measures the UE Rx-Tx measurements (and optionally the DL PRS RSRP of the received signal) using assistance data received from a positioning server, and the TRPs measure the gNB Rx-Tx measurements (and optionally the UL SRS RSRP of the received signal) using assistance data received from a positioning server. The measurements are used to determine the RTT at the positioning server, which is used to estimate the UE's position.

[0080] In the enhanced cell identifier (E-CID) positioning method, the UE's position is estimated using knowledge of its serving ng-eNB, gNB, and cell, and is based on LTE signals. Information about the serving ng-eNB, gNB, and cell may be acquired through paging, registration, or other methods. NR E-CID positioning refers to techniques that use additional UE measurements and / or NR radio resource and other measurements to improve the UE position estimate using NR signals. Although NR E-CID positioning can utilize some of the same measurements as the measurement control system in the RRC protocol, it is generally not expected that the UE will make additional measurements solely for positioning; for example, the positioning procedure does not provide measurement configuration or measurement control messages, and the UE reports the measurements it has available without taking additional measurement actions.

[0081] The UL TDOA positioning method utilizes the UL TDOA (and optionally UL SRS-RSRP) of the uplink signal transmitted from the UE at multiple reception points (RPs). The RP uses assistance data received from the positioning server to measure the UL TDOA (and optionally UL SRS-RSRP) of the received signal, and the resulting measurements are used together with other configuration information to estimate the UE's position.

[0082] The UL AoA positioning method utilizes the azimuth and zenith of arrival of uplink signals sent from the UE measured at multiple positioning nodes (RPs). The RP uses assistance data received from a positioning server to measure the A-AoA and Z-AoA of the received signal, and the resulting measurements, along with other configuration information, are used to estimate the UE's position.

[0083] RAT-independent positioning techniques can also be implemented, including [TS38.305]:

[0084] Network-assisted Global Navigation Satellite System (GNSS) methods: These methods utilize UEs equipped with radio receivers capable of receiving GNSS signals. In the 3GPP specifications, the term GNSS includes both global and regional / augmented navigation satellite systems. Examples of global navigation satellite systems include the Global Positioning System (GPS), modernized GPS, Galileo, GLONASS, and the BeiDou Navigation Satellite System (BDS). Regional navigation satellite systems include the Quasi-Zenith Satellite System (QZSS), while some augmentation systems are classified as a general term for Space-Based Augmentation Systems (SBAS) and provide regional augmentation services. In this concept, different GNSS (e.g., GPS, Galileo, etc.) can be used alone or in combination to determine the position of the UE.

[0085] Barometric pressure sensor positioning: The barometric pressure sensor method uses a barometric pressure sensor to determine the vertical component of the UE's position. The UE measures the air pressure (optionally with the help of assistance data) to calculate the vertical component of its position, or sends the measurement to a positioning server for position calculation. This method can be combined with other positioning methods to determine the UE's 3D position.

[0086] Wireless Local Area Network (WLAN) Positioning: WLAN positioning methods use WLAN measurements (access point (AP) identifiers and optionally other measurements) and a database to determine the UE's location. The UE measures the signals received from the WLAN access points (optionally with the aid of assistance data) and sends the measurements to a positioning server for position calculation. Using the measurements and a reference database, the UE's position is calculated. Alternatively, the UE determines its position using WLAN measurements and optional WLAN AP assistance data provided by the positioning server.

[0087] Bluetooth positioning: The Bluetooth positioning method uses Bluetooth measurements (beacon identifiers and optional other measurements) to determine the UE's location. The UE measures the signals received from Bluetooth beacons. Using the measurements and a reference database, the UE's position is calculated. The Bluetooth method can be combined with other positioning methods (e.g., WLAN) to improve the UE's positioning accuracy.

[0088] TBS positioning: TBS consists of a network of ground-based transmitters that broadcast signals used only for positioning purposes. Current types of TBS positioning signals are MBS (Metropolitan Beacon System) signals and PRS (Technical Specification (TS) 36.211 [4]). The UE measures the received TBS signals (optionally with the help of assistance data) to calculate its position, or sends the measurements to a positioning server for use in position calculation.

[0089] Motion sensor positioning: The motion sensor method uses various sensors, such as accelerometers, gyroscopes, and magnetometers, to calculate the UE's displacement. The UE estimates relative displacement based on a reference position and / or reference time. The UE sends a report containing the determined relative displacement, which can be used to determine the absolute position. This method can be used in conjunction with other positioning methods for hybrid positioning.

[0090] Figure 4a and Figure 4b Illustrated is a portion of an LPP RequestLocationInformation message 400. The RequestLocationInformation message 400 body in an LPP message may be used by a location server to request positioning measurements or a location estimate from a target device.

[0091] Figure 5a and Figure 5b Illustrated is a portion of an LPP ProvideLocationInformation message 500. The ProvideLocationInformation message 500 body in an LPP message may be used by a target device to provide positioning measurements or a position estimate to a location server.

[0092] For RAT-related positioning measurements, the following table 4 shows different DL measurements used for supported RAT-related positioning technologies, including DL PRS-RSRP, DL RSTD and UE Rx-Tx time difference. For example, the following measurement configuration is specified [TS38.215]:

[0093] • 4 pairs of DL RSTD measurements may be performed for each pair of cells. Each measurement is performed between a different DL PRS resource pair / resource set with a single reference timing.

[0094] • 8 DL PRS RSRP measurements may be performed on different DL PRS resources from the same cell.

[0095] Table 4: DL measurements required for DL-based positioning methods [TS 38.215]

[0096]

[0097]

[0098]

[0099] Figure 6 System 600 for machine learning-based RAN intelligence is illustrated. In system 600, data collection is a function that provides input data to model training and model inference functions. AI / ML algorithm-specific data preparation (e.g., data preprocessing and cleaning, formatting, and transformation) may not be performed in the data collection function. Examples of input data may include measurements from UEs or different network entities, feedback from participants, and output from AI / ML models.

[0100] ○ Training data: Data required as input to the AI / ML model training function.

[0101] ○ Inference data: Data required as input to the inference function of the AI / ML model.

[0102] Model training is a function that performs ML model training, validation, and testing as part of the model testing process, generating model performance metrics. The model training function is also responsible for data preparation (e.g., data preprocessing and cleaning, formatting, and transformation) based on the training data obtained by the data collection function.

[0103] Model deployment / update: Used to initially deploy trained, validated, and tested AI / ML models to the model inference function, or to deliver updated models to the model inference function.

[0104] Model inference is a function that provides AI / ML model inference output (e.g., predictions or decisions). If necessary, the model inference function is also responsible for data preparation (e.g., data preprocessing and cleaning, formatting, and transformation) based on the inference data obtained by the data collection function.

[0105] Output: The inference output of the AI / ML model generated by the model inference function.

[0106] Model performance feedback: If certain information derived from the model inference function is applicable to improving the AI / ML model trained in the model training function, then this feedback is applied. At the model inference function, feedback from participants or other network entities (via the data collection function) may be required to create model performance feedback.

[0107] An actor is a function that receives the output from a model inference function and triggers or performs corresponding actions. An actor can trigger actions directed to other entities or itself.

[0108] Feedback: Information that can be used to derive training or inference data or performance feedback.

[0109] Therefore, the present disclosure provides solutions that support machine learning for positioning. For example, AI-based direct positioning and AI-assisted positioning methods can be used to improve the location accuracy performance of UEs. In scenarios for direct AI / ML positioning, AI / ML models can use measurement and environmental data using techniques such as fingerprinting to obtain enhanced location accuracy. The present disclosure describes techniques for configuring direct AI / ML positioning assistance data and defining measurements to perform AI / ML direct positioning. In addition, the present disclosure describes techniques for configuring reporting standards for devices, nodes, and / or entities performing AI / ML direct positioning measurements.

[0110] With respect to aspects of the present disclosure, implementations are described for: configuring a target UE, positioning reference unit (PRU) UE, sidelink (SL) UE, or NG-RAN node to perform direct AI / ML positioning measurements based on positioning reference signal transmission to generate a training data set; enabling the target UE, PRU UE, SL UE, or NG-RAN node to perform requested measurements for different scenarios, which may form part of a fingerprint based on the environment topology and AI / ML measurement parameters; enabling the target UE, PRU UE, NG-RAN node, configuration entity, SL UE, and / or location server to indicate AI / ML positioning assistance data or measurement error causes; enabling a reporting configuration framework for the target UE, PRU UE, SL UE, and / or NG-RAN node to receive desired direct AI / ML positioning measurements, such as fingerprint identification information; implementing multiple common reporting standards for AI / ML positioning measurements; and enabling configuration and reporting of assistance information for accurately reporting AI / ML positioning measurements.

[0111] Some notes about the implementations described in this disclosure: different implementations may be combinable with each other in various ways; positioning-related reference signals may refer to reference signals used for positioning processes and / or purposes to estimate the position of a target UE, such as PRS, signals based on existing reference signals (such as channel state information (CSI) reference signals (RS) (CSI-RS) or SRS), etc.; a target UE may refer to a device and / or entity whose position and / or location is to be determined; the term "PRS" may refer to any signal, such as a reference signal, that may or may not be used primarily for positioning; a target UE may be referred to as a UE of interest, whose position (e.g., absolute and / or relative position) is to be obtained by the network and / or by the UE itself; the terms AI and ML may be used interchangeably to refer to intelligent software components or systems, and AI may represent a subset and / or implementation of ML; references to device position and / or positioning information may refer to 2D / 3D absolute position, relative position relative to another node and / or entity, distance-based ranging, direction-based ranging, and combinations thereof.

[0112] The implementations disclosed herein support configurations for direct AI / ML positioning measurements and processing. For example, fingerprinting is described, such as where inference AI / ML models can be deployed at different entities. Examples of such implementations include UE-based positioning using UE-side ML patterns, UE-assisted and / or LMF-based positioning using LMF-side ML models, NG-RAN node-assisted positioning using LMF-side ML models, and the like.

[0113] In implementations including UE-based positioning, the target UE may request a training data set from various data sources based on DL and SL reference signals (e.g., DL-PRS, SL-PRS, etc.), including measurements performed and collected internally within the target UE, other UEs, location servers, NG-RAN nodes, PRUs, network operations, administration, and maintenance (OAM), trace collection entities (TCEs), and / or combinations thereof. The training data set may include reference points and / or reference locations, and fingerprint information and / or other positioning measurements may be sampled, measured, and / or correlated. Various network entities and / or nodes may be enabled using the following procedures to enable configuration:

[0114] The target UE, anchor UE, and / or PRU UE may send a request and receive a response including a configuration for downlink (DL) direct AI / ML positioning assistance (e.g., configuration) data, such as fingerprinting with respect to a location server;

[0115] The target UE may send a request and receive training data (e.g., instead of a configuration to perform measurements) or instructions to obtain training data from a second node (e.g., a location server, an NG-RAN node, a positioning reference element, a network OAM, a TCE, or a combination thereof);

[0116] ● Another node (e.g., a location server, an NG-RAN node, and / or another node)

[0117] A request may be sent to the target UE to receive training data that has been collected and / or measured by the target UE;

[0118] The location server may receive DL direct AI / ML positioning assistance (e.g., configuration) data, such as data for a radio frequency (RF) fingerprinting request, from multiple UEs including a target UE and a PRU UE, and the location server may provide a configuration response;

[0119] One or more UEs including anchor UEs and / or PRU UEs may receive a request for SL direct AI / ML positioning assistance data (e.g., configuration and / or RF fingerprinting),

[0120] And can provide appropriate configuration response.

[0121] In implementations including UE-assisted positioning, the location server may request multiple training data sets from various data sources based on DL and SL reference signals (e.g., DL-PRS, SL-PRS), including measurements collected internally within the location server, multiple UEs, other neighboring location servers, NG-RAN nodes, positioning reference units, and combinations thereof. These data sets may contain reference points and / or reference locations where fingerprint information may be sampled, measured, and / or associated. Various network entities and / or nodes may be enabled with the following functionality to enable such an implementation:

[0122] ●The location server can send downlink (DL) or sidelink (SL) direct AI / ML positioning assistance (configuration), such as fingerprint recognition data.

[0123] • At least one or more UEs including a target UE or a PRU UE may receive a response / configuration for DL ​​and / or SL AI / ML direct positioning configuration, eg, fingerprinting with respect to a location server.

[0124] Figure 7An example scenario 700 supporting machine learning for positioning according to aspects of the present disclosure is illustrated. For example, the scenario 700 includes a representation of DL and SL direct AI / ML configuration mechanisms within a rectangular environment 702 (e.g., an indoor factory floor environment) having a length L and a width W, including a reference point and 18 gNBs / TRPs separated by a gNB / TRP distance D, as described above.

[0125] In scenario 700, DL LTE Positioning Protocol (LPP) signaling 704 from a location server 706 to a target UE 708 and / or an anchor / PRU UE 710 can be utilized to convey multiple direct AI / ML positioning configurations, for example, using the LP ProvideAssistanceData message or the new LPP ProvideMLAssistanceData message, while the Sidelink Positioning Protocol (SLPP) or a new positioning protocol associated with the exchange of SL positioning messages can be used for signaling 712 to provide multiple direct AI / ML positioning configurations from the anchor / PRU UE 710, for example, using the SLPP ProvideAssistanceData message. Alternatively or additionally, the SL positioning server UE can provide direct AI / ML positioning configurations via SLPP or the like. Furthermore, the target UE 708 can configure surrounding UEs, PRU UEs, and / or SL positioning server UEs to perform direct AI / ML positioning measurements, such as fingerprinting based on reference locations / points. For illustrative purposes, direct signaling from the location server 706 is shown. However, such signaling can be transparently routed to the UE via the serving gNB.

[0126] In implementation, an anchor / PRU UE 710 may configure other anchor / PRU UEs 710 to perform AI / ML positioning measurements. In addition, the LMF may configure a first set of anchor / PRU UEs 710 to send SLPRS to a second set of anchor / PRU UEs 710.

[0127] In an implementation such as that shown in scenario 700, the target UE 708, the anchor / PRU UE 710 and / or the location server 706 may request multiple DL or SL direct AI / ML positioning assistance data, for example using an LPP RequestAssistanceData message. Alternatively or additionally, the target UE 708 may request multiple SL direct AI / ML positioning assistance data from the anchor / PRU UE 710 and / or the SL positioning server UE, such as using an SLPP and / or SL positioning protocol message RequestAssistanceData.

[0128] In implementations including NG-RAN assisted positioning, the location server may request multiple training data sets from various data sources based on UL reference signals (e.g., SRS for positioning), including neighboring gNBs, TRPs, NG-RAN nodes, PRU TRPs, CUs, DUs, and / or combinations thereof. The training data sets may include reference points and / or reference positions corresponding to NG-RAN nodes, CU positions, and / or DU positions, and fingerprint information may be sampled, measured, and / or correlated. In implementations, various network entities and / or nodes may have the following procedures:

[0129] ● NG-RAN nodes including gNB and / or TRP may send UL AI / ML direct positioning configuration to one or more UEs upon request from the location server (e.g.

[0130] fingerprint recognition).

[0131] • UEs including target UEs may receive a response and / or configuration for UL AI / ML direct positioning configuration (eg, fingerprinting) with a location server.

[0132] ●The location server can send UL AI / ML direct positioning assistance, such as fingerprinting data request, to one or more NG-RAN nodes (including gNB, TRP, CU, DU, PRU and / or their combination) and then receive a configuration response.

[0133] Figure 8 An example scenario 800 for supporting machine learning for positioning according to aspects of the present disclosure is illustrated. In the scenario 800, at step 802, a location server 804 triggers a configuration request to an NG-RAN node 806 (e.g., a serving gNB and / or a neighboring gNB). For example, step 806 involves NRPPa with UL direct AI / ML positioning configuration, such as fingerprinting of N reference locations.

[0134] At step 808, the NG-RAN node 806 forwards the UL-RS configuration (e.g., SRS for positioning configuration) to the target UE 810 and / or PRU / anchor UE 812 to enable UL transmission and subsequent UL direct AI / ML measurements at the gNB / TRP. For example, step 808 may involve RRC with UL direct AI / ML positioning configuration, such as fingerprinting of N reference locations.

[0135] In an implementation, the location server 804 may directly forward the UL-RS configuration via DL LPP signaling for use in performing UL transmission and subsequent UL direct AI / ML measurements at the NG-RAN node 806 (e.g., gNB / TRP). In an alternative implementation, the NG-RAN node 806 may be implemented as a CU and a DU, where the location server 804 may signal the CU and the DU may perform UL-RS measurements at different distributed reference locations and / or points (e.g., in an indoor factory scenario as described above). For example, at step 814, the location server 804 may send a UL direct AI / ML positioning configuration, such as fingerprinting of N reference locations.

[0136] In an implementation, the PRU / anchor UE 812 location may be utilized along with the gNB / TRP ground truth reference location to construct a fingerprint comprising the ground truth reference location. In such an implementation, the PRU / anchor UE 812 may signal the 2D and / or 3D location at which the SRS is being sent to a configuration entity, such as the NG-RAN node 806 and / or the location server 804. In scenarios involving signaling location information to the location server 804, LPP signaling (e.g., ProvideLocationInformation message) may be used, and / or in scenarios involving signaling location information to the NG-RAN node 806, RRC signaling (e.g., LocationMeasurementIndication) may be used.

[0137] Alternatively or additionally, the configuration entity may request the PRU / anchor UE 812 to transmit UL SRS or SL PRS at certain predefined locations. These predefined locations may be included in the direct AI / ML configuration along with the positioning reference signal configuration.

[0138] In an implementation, one type of NG-RAN node (e.g., PRU gNB / TRP) may also send SRS to other NG-RAN nodes, UEs and / or devices. An NG-RAN node receiving the SRS may perform direct AI / ML positioning measurements for the purpose of direct AI / ML position estimation. Therefore, NG-RAN nodes that are capable of sending and receiving SRS (such as PRU TRP) may support this type of measurement. In an implementation, new reference signals may also be supported between NG-RAN nodes for the purpose of direct AI / ML position estimation, and the new reference signals may be sent, such as via an Xn interface and / or via another wireless transmission medium. In an implementation, the above configuration methods may be combined, such as to enable DL, SL and / or UL direct AI / ML positioning measurements to be utilized individually or in combination.

[0139] In an implementation, a configuration such as the signaling described above may be used to enable direct AI / ML position estimation, for example using a fingerprinting approach. For example, the configuration content is considered in light of the above implementation and is described below.

[0140] In a UE-based positioning scenario (such as utilizing a UE-side model), the target UE may create multiple fingerprint training data sets based on measurements performed, data, received measurements, and / or data from other network entities (e.g., a location server, other UEs, PRUs, etc.). The target UE may initiate a request to the location server for a configuration related to measurements of DL or SL direct AI / ML measurements (e.g., fingerprint measurements), and / or request a transmission configuration of UL-RS related to direct AI / ML positioning. Example configurations are discussed below.

[0141] Figure 9 Message 900 supporting machine learning for positioning according to aspects of the present disclosure is illustrated. For example, message 900 represents an NR-Direct-AI-ML-AssistanceData information element that can be used by a target device to request direct AI / ML positioning assistance data from a location server and / or a configuration entity (e.g., a SL positioning server UE, an anchor UE, etc.).

[0142] Table 5 below provides example field descriptions for message 900 .

[0143] Table 5

[0144]

[0145]

[0146] In implementations, message 900 may include direct AI / ML positioning assistance data, assisted AI / ML positioning, or a combination thereof. Assisted AI / ML positioning measurements may be defined as measurements enhanced and / or optimized using AI / ML models, for example, RAT-related measurements such as RSTD, relative time of arrival (RTOA), RSRP, RSRPP, Rx-Tx, AoA, AoD, time difference, etc.

[0147] In implementation, with respect to UE-based positioning as described above, the target UE may provide an index or list of ground truth reference positions, such as absolute and / or relative positions.Example ground truth reference positions are defined as follows.

[0148] In implementations such as those involving UE-based positioning (e.g., UE-assisted positioning using a UE-side model and / or using an LMF-side model), the location server may send direct AI / ML positioning assistance data. Alternatively or additionally, the location server may send assisted AI / ML positioning assistance data for performing direct AI / ML and / or assisted AI / ML measurements to one or more target devices, PRU UEs, anchor UEs, SL positioning server UEs, etc., to be used as training data input. For example, the target UE may receive from the location server a plurality of configurations related to measurements of DL or SL direct AI / ML measurements, such as fingerprint measurements and / or transmission configurations of UL-RS related to direct AI / ML positioning.

[0149] Figure 10a and Figure 10b 1000, for example, a NR-Direct-AI-ML-ProvideAssistanceData message that supports machine learning for positioning.

[0150] Table 6 below provides example field descriptions for message 1000.

[0151] Table 6

[0152]

[0153]

[0154]

[0155] In implementation, the above DL-PRS configuration information may be used to allow a device (e.g., target UE, PRU UE, SLUE, etc.) to perform direct AI / ML measurements at each configured ground truth reference location, as signaled above. This process may be performed during an offline phase, and the measurements and corresponding locations may be signaled to a location server or stored in a single UE or multiple UEs.

[0156] In implementations involving NG-RAN assisted positioning (e.g., utilizing the LMF side model), a location server may send one or more requests for multiple direct AI / ML positioning assistance data to multiple gNBs and / or TRPs based on the available SRS and / or other UL-PRS configurations. One or more gNBs and / or TRPs may determine an SRS configuration for each target UE to use for performing SRS transmission-dependent AI / ML positioning, which may be configured per carrier. In implementations, the SRS configuration may be broadcast to multiple UEs for use in multiple cells, within predefined positioning system information areas, within areas with associated validity in terms of time and / or area, and combinations thereof.

[0157] In an implementation, the gNB and / or TRP may configure the UE to perform SRS for positioning transmission via RRC signaling using, for example, an RRCReconfiguration message. The gNB and / or TRP may configure the UE to perform SRS for positioning transmission in order to perform direct AI / ML positioning or assisted AI / ML positioning.

[0158] In an implementation, a target UE may confirm reception of the SRS for positioning configuration to perform direct AI / ML positioning and / or assisted AI / ML positioning measurements, as well as other non-AI / ML timing or angle-based measurements. A location server (e.g., LMF) may request one or more gNBs and / or TRPs to activate the SRS for positioning configuration for transmission by the target UE. The gNB and / or TRP may activate SRS transmission to the target UE by sending a DL MAC Control Element (CE) activation command to the target UE. The location server may also deactivate SRS transmission via the gNB, and the gNB may send the deactivation command using, for example, a DL MAC CE.

[0159] In an implementation, a location server may receive multiple available SRS or UL-PRS configurations for performing UL direct AI / ML positioning. In an example scenario, the gNB may derive SRS and / or UL-PRS fingerprints for position estimation of multiple UEs.

[0160] In an implementation, low-layer signaling (e.g., RRC signaling, MAC CE, downlink control information (DCI) signaling, and combinations thereof) with respect to LPP may be used to convey direct AI / ML or assisted AI / ML configurations. In at least one implementation, LPP and / or RRC signaling may be used to add, modify, remove, update, activate, and / or deactivate one or more UE direct AI / ML positioning configurations.

[0161] Implementations enable direct AI / ML measurements and processing. For example, based on an AI / ML approach, methods for performing direct AI / ML measurements are presented. In at least one implementation, fingerprint measurements may rely on received signal strength (RSS) measurements, including RSRP, reference signal received quality (RSRQ), RSSI, or a combination thereof. In implementations, fingerprint measurements may include a combination of RSS, timing-based, and angle-based measurements to obtain a position estimate for the UE.

[0162] In implementation, the target UE, PRU UE, SL UE, etc. may be configured to measure the following measurements to construct an RF fingerprint applicable to:

[0163] RAT-related measurements:

[0164] ○DL / SL RSTD (DL-based measurement or SL-based measurement)

[0165] ○DL / SL PRS Time of Arrival (TOA) (DL-based measurement or SL-based measurement)

[0166] ○DL / SL PRS RSRP (DL-based measurement or SL-based measurement)

[0167] ○DL / SL PRS RSRPP (DL-based measurement or SL-based measurement)

[0168] UE Rx-Tx time difference (DL-based measurement or SL-based measurement)

[0169] ○ Synchronization Signal (SS)-RSRP (RSRP for RRM), SS-RSRQ (for RRM), CSI-RSRP (for RRM), CSI-RSRQ (for RRM), SS-RSRP (for RRM) (DL-based measurement)

[0170] ○DL / SL carrier phase measurement (DL-based measurement or SL-based measurement)

[0171] ○DL / SL carrier phase difference measurement (DL-based measurement or SL-based measurement)

[0172] DL E-CID

[0173] LTE E-CID

[0174] ○SL physical sidelink shared channel (PSSCH) RSRP,

[0175] DL PRS RSSI (Received Signal Strength Indicator)

[0176] LTE Observed Time Difference of Arrival (OTDOA) measurement

[0177] RAT-independent measurements

[0178] o A-GNSS measurements, including public assistance data that may be applicable to any GNSS constellation (e.g., Galileo, GPS, GLONASS, etc.), general assistance data for a specific GNSS constellation, or GNSS assistance data for a periodic provision of GNSS control information to the UE / device.

[0179] Bluetooth RSS measurements, including RSSI

[0180] ○WLAN (WiFi) measurements, including RSSI and RTT information

[0181] ○IMU sensor measurements, including gyroscopes, accelerometers, etc.

[0182] ○Air pressure sensor measurement

[0183] In implementation, NG-RAN nodes, gNB / TRP, CU-DU, etc. may be configured to measure the following measurements applicable to:

[0184] RAT-related measurements

[0185] UL-RTOA (measurement based on UL)

[0186] UL SRS RSRP (based on UL measurement)

[0187] UL SRS RSRPP (measurement based on UL)

[0188] gNB Rx-Tx time difference measurement (UL-based measurement)

[0189] UL AoA (UL-based measurement)

[0190] UL carrier phase measurement (UL-based measurement)

[0191] UL carrier phase difference measurement (UL-based measurement)

[0192] ○UL NR E-CID

[0193] LTE E-CID

[0194] The above measurement examples can form part of multiple fingerprint measurements, including DL or UL direct AI / ML positioning measurements at one or more ground truth locations. Using RAT-dependent and RAT-independent methods can help derive a hybrid fingerprint to improve the accuracy of direct AI / ML positioning methods.

[0195] In implementations, a location server (e.g., LMF) may supply direct AI / ML and / or assisted AI / ML configurations for measurements, such as those based on AI and / or machine learning models. These models may include, but are not limited to, any one or more of the following combinations:

[0196] Supervised learning methods:

[0197] k-NN (nearest neighbor) clustering

[0198] ■The model classifies fingerprints based on the Euclidean distance between adjacent training data points and determines the K closest neighbors to the input fingerprint.

[0199] ○Support Vector Machine (SVM)

[0200] ■The model is based on edge computing, where the input fingerprint data is plotted in an n-dimensional space, where n-1 hyperplanes are drawn to divide the training data into n classes such that the distance between each class and the hyperplane is maximized.

[0201] Decision Tree

[0202] Classification or regression problems related to fingerprint matching can be solved using a decision tree-like structure. Rules are used to split the training data into multiple labels, where the label is predicted for any new fingerprint data point by the decision tree.

[0203] Random Forest

[0204] The model is an ensemble of multiple decision trees where the result of each tree provides the fingerprint classification, or in another implementation, the average prediction of all decision trees is in the output. This helps overcome the overfitting problem experienced by standalone decision trees.

[0205] Artificial Neural Network (ANN)

[0206] Based on the back-propagation learning algorithm, the input dataset of fingerprint data is converted into the output of the final position estimate using nonlinear transfer functions within the intermediate units / nodes including the hidden layer. Such a model can be robust to noisy or interference-limited fingerprint data.

[0207] Unsupervised methods:

[0208] K-means

[0209] ■The model segments the fingerprint into K unique and non-overlapping clusters or groups that represent specific location points.

[0210] ○ Gaussian Mixture Model (GMM)

[0211] ■GMM is a probabilistic model that can be used to estimate the distribution of RF fingerprints in different locations, including ground truth reference locations as well as unknown locations. GMM can be trained using collected RF fingerprints and can then be used to determine the most likely location for a given set of RF fingerprints.

[0212] Both supervised and unsupervised methods:

[0213] ○ Bayesian Network (BN)

[0214] ■BN can be used to model the probabilistic relationship between the RF fingerprint for a given location and environmental factors such as radio channel parameters (e.g., channel state information (CSI), path loss, fading parameters). The BN can be trained using a training set of RF fingerprint and environmental data to perform RF fingerprint positioning.

[0215] Study

[0216] ○ Reinforcement Learning

[0217] ■A learning algorithm based on trial and error, where decisions are based on so-called “rewards” or “penalties,” where correct decisions are rewarded and incorrect decisions are penalized to improve model performance.

[0218] Deep Learning

[0219] ■ Based on ANN, it uses iterative weight adjustment techniques between a pair of neurons / nodes that are trained using a large set of fingerprint data collected from the environment.

[0220] ○ Transfer learning

[0221] Leveraging the model's ability to learn new features and topics based on its own system knowledge, which allows for minimal changes to existing trained models. This can provide a scalable solution using direct AI / ML localization (e.g., fingerprinting) to avoid the significant overhead of collecting site fingerprints during the initial site survey (e.g., during an offline phase).

[0222] In implementation, the location server may explicitly indicate the ML model to the target UE and / or PRU UE using UE-specific LPP / SLPP signaling (e.g., SLP / LP ProvideAssistanceData message). In implementation, the model may be indicated to multiple UEs using positioning system information broadcast messages (e.g., new and / or existing posSIBs).

[0223] In implementation, the location server and / or configuration entity may receive a request for direct AI / ML positioning or assisted AI / ML positioning assistance data from a target UE, PRU UE, and / or SL UE, and an indication of the above-mentioned model for which measurements are to be used as input data.

[0224] In an implementation, the ML model used to initiate a direct AI / ML or assisted AI / ML positioning session may be indicated via applicable UE capability signaling (e.g., LPP ProvideCapabilites message), which may be based on a sought request from a location server and / or configuration entity, such as an LPP RequestCapabilities message.

[0225] In implementation, PRS RSSI measurements may be defined for AI / ML positioning purposes, including both direct and assisted techniques. In implementation, PRS RSSI may be applicable to non-AI / ML positioning techniques. Furthermore, these measurements may be performed in the RRC_CONNECTED, RRC_INACTIVE, and RRC_IDLE states. The PRS RSSI for downlink, sidelink, and uplink may be defined as follows in Table 7.

[0226] Table 7: DL, SL, and UL PRS RSSI measurement definitions

[0227]

[0228]

[0229]

[0230] In implementation, according to the measurement definitions presented in Table 7, DL, SL, and UL PRS RSSI may include features or signatures that can be used as fingerprints to enable direct AI / ML positioning. In addition, PRS or SRS TOA measurements may also be defined as additional features or signatures for AI / ML positioning purposes, including both direct and assisted techniques. In implementation, PRS TOA may be applicable to non-AI / ML positioning techniques. In addition, these measurements may be performed in RRC_CONNECTED, RRC_INACTIVE, and RRC_IDLE states. PRS TOA for downlink, sidelink, and uplink (SRS) is defined as shown in Table 8 below.

[0231] Table 8: DL, SL, and UL PRS TOA measurement definitions

[0232]

[0233]

[0234]

[0235] In implementations, the target UE and / or PRU UE may be configured to measure DL PRS and / or SL PRS for AI / ML positioning, such as in one or more of the following scenarios:

[0236] • Configured with a Measurement Gap (MG) associated with a Gap Pattern ID, a Measurement Gap Length (MGL) and a Measurement Gap Repetition Period (MGRP) or a combination thereof. The MG may be pre-configured with an activation command or a deactivation command.

[0237] • If the DL PRS is located within the active DL bandwidth part (BWP) with the same digital technology as the active DL BWP, it is configured with no or beyond measurement gaps.

[0238] Configured with a training measurement gap (T-MG), where a gap pattern ID, measurement gap length (MGL) and measurement gap repetition period (T-MGRP) are predefined, along with a T-MG start time, T-MG duration and T-MG end time, a flag indicating whether the measurement is based on offline or online training, or a combination thereof. In other implementations, a duration given by a measurement window or timer expiration can be used to indicate the start and end of the duration of measurements for online or offline training of an AI / ML positioning model.

[0239] In implementation, for each received DL / SL PRS, the UE may be configured with a priority configuration of PRS resources for performing AI / ML positioning and non-AI / ML positioning measurements. These resources may form a subset of resources, which may form part of the same or different PRS resource sets. The priority signaled to the UE / device may indicate the priority of performing measurements, which may additionally or alternatively construct a training data set for performing measurements for non-AI / ML positioning.

[0240] Implementations also provide for providing assistance data and / or measurement error causes. For example, the UE may indicate to the network and / or configuration entity that one or more measurements associated with the ground truth location and / or fingerprint have an associated error cause. For example, the error cause may be that the PRS configuration was not received or that the PRS configuration lacks configuration parameters. For example, using the above implementation, LPP and / or SLPP signaling may be used to indicate the error cause to the network. Furthermore, the error cause may be UE-initiated, such as initiated on the UE side. In implementations, the error cause may be an indication to the UE initiated by a location server.

[0241] Figure 11Illustrated is a message 1100 that supports machine learning for positioning according to aspects of the present disclosure. For example, message 1100 represents an IE that shows error causes supported by a location server and / or configuration entity, such as error causes that can be communicated via LPP. For example, message 1100 represents the NR-AI-ML-LocationServerErrorCauses IE, which a location server can use to provide AI / ML assistance data error causes to a target device. Message 1100 can also be used to provide a SL configuration entity with such error causes to a target UE and / or device.

[0242] Figure 12 Illustrated is a message 1200 that supports machine learning for positioning according to aspects of the present disclosure. For example, message 1200 presents error causes supported by the target UE, which can be communicated via LPP. For example, message 1200 represents the NR-AI-ML-TargetDeviceErrorCauses IE, which the target UE can use to provide NR direct AI / ML or assisted AI / ML measurement error causes to the location server. Such an implementation can be applicable to SL target UEs and / or devices that provide the above error causes to the SL configuration entity.

[0243] In implementation, the NG-RAN node may signal the NR-AI-ML-NG-RANnodeErrorCauses IE with one or more combinations of the parameters included in the above-mentioned NR-AI-ML-TargetDeviceErrorCauses IE. The error causes derived at the NG-RAN node (e.g., gNB and / or TRP) may be signaled to the location server via the NRPPa interface (e.g., using an Error Indication message). Additionally, the following error causes may be signaled from the NG-RAN node side:

[0244] • The target UE or PRU UE has moved to another cell.

[0245] The requested AI / ML positioning measurements could not be provided on time.

[0246] ●AI / ML training models are invalid and need to be retrained.

[0247] The AI / ML inference model is invalid and needs to be re-acquired.

[0248] In an implementation, positioning measurements including the definitions contained in Tables 7 and 8 may have associated quality indicators that indicate the following:

[0249] • The quality of the performed measurements based on timing and RSS parameters.

[0250] • The quality of the measurements performed relative to similar measurements performed in surrounding ground truth reference locations.

[0251] The implementations described herein also provide a reporting configuration process. For example, an implementation of reporting configuration that enables direct AI / ML positioning measurement and reporting (e.g., fingerprinting) is described for the following scenario, where an inference AI / ML model can be deployed at the following entities to perform positioning:

[0252] UE-based positioning using UE-side models;

[0253] UE-assisted positioning / LMF-based positioning using LMF side model;

[0254] ● Utilize NG-RAN node assisted positioning on the LMF side model.

[0255] The above scenarios are presented as examples only, and the corresponding details can be extended beyond these example scenarios.

[0256] In the scenario of UE-based positioning, the target UE can request training data sets from various data sources based on DL and SL reference signals (e.g., DL-PRS, SL-PRS, etc.), including measurements performed and collected internally within the target UE, at other UEs, at a location server, at an NG-RAN node, at a PRU, at OAM, at a TCE, and / or at a combination thereof. In addition, training can be performed on the UE side, and the calculated position estimate can be calculated by the target UE device. In such an implementation, a reporting configuration including reporting criteria and / or assistance information and associated measurement reports can be initiated from the target UE.

[0257] Various network entities, UEs, and / or nodes may be enabled using the following procedures to enable reporting configurations, such as in scenarios where AI / ML model training is performed on the UE side.

[0258] Figure 13 A scenario 1300 is illustrated for supporting machine learning for positioning according to aspects of the present disclosure. For example, the scenario 1300 includes a scenario 1300a in which UE-side training can be performed using target UE reasoning with other UEs, and a scenario 1300b in which UE-side training can be performed using target UE reasoning with a network entity.

[0259] In scenario 1300, at 1302, a target UE 1304 may request a plurality of direct AI / ML positioning measurements based on defined reporting criteria. For example, in scenario 1300a of signaling to the UE, at 1304a, the SL Positioning Protocol (SLPP) and / or other defined positioning protocols may be employed by the target UE 104a to request a direct AI / ML positioning report including measurement reporting criteria and / or assistance information from the UE 104b. At 1300b, LPP signaling may be employed at 1304b for a network entity 102, such as a location server, to request a direct AI / ML positioning report including measurement reporting criteria and / or assistance information. In scenarios involving NG-RAN nodes, RRC and / or related UL signaling, such as UL MAC CE, may be utilized.

[0260] In addition to the scenario 1300, the corresponding network entity 102 and / or UE 104 can provide a response 1306 to the request 1304 to provide measurements related to the reporting criteria and / or assistance information indicated in the request 1304. For example, UE 104b can respond via SLPP at 1306a, and the network entity can respond via LPP and / or RRC at 1306b. In the scenario where the network entity 102 reports the measurements, the location server can receive measurement reports from other UEs (e.g., PRU UEs and / or other UEs) based on solicited or unsolicited requests and report such measurements to the target UE 104a.

[0261] At 1308, the target UE 104a may construct a training dataset based on the measurement reports from the different sources and perform training of the AI / ML model. In an implementation, the target UE 104a may utilize the trained ML model and perform inference based on the new measurement data by repeating steps 1304a, 1304b and 1306a, 1306b to obtain new measurement information for processing by the trained ML model. In an implementation, the training and inference datasets may be requested in a single step to avoid repeating steps 1304 and 1306 to train the ML model and then perform inference.

[0262] In implementations, ML model training can be performed on the network side (e.g., at a location server or NG-RAN node (e.g., gNB)) and / or at other UEs / devices (e.g., anchor UEs, PRU UEs, SL UEs, etc.). In such implementations, network entities and / or UEs / devices can request multiple reporting criteria so that desired measurements and assistance information are reported in a timely and accurate manner. These datasets can contain fingerprint information or other reference points or reference locations where positioning measurements are sampled or measured or associated.

[0263] Figure 14a and Figure 14b Scenario 1400 for supporting machine learning for positioning according to aspects of the present disclosure is illustrated. For example, scenario 1400 includes scenarios 1400a and 1400c, in which UE-side training is performed using target UE inference, and scenario 1400b, in which network-side training is performed using target UE inference.

[0264] In the scenario 1400, at 1402, the corresponding network entity 102 and / or UE 104 may request a plurality of direct AI / ML positioning measurements from the target UE 104a based on certain defined reporting criteria. As for the signaling transmission to the UE 104, SLPP and / or a newly defined positioning protocol may be adopted, while in the case of a network entity 102 such as a location server, LPP signaling may be adopted, and in the scenario for an NG-RAN node, RRC or any related DL signaling, such as DL MACCE, may be adopted.

[0265] At 1404, the target UE 104a may respond to the received request and provide the requested measurements according to the reporting criteria and / or assistance information. For the network entity 102, the location server may receive measurement reports from other UEs (e.g., PRU UEs, other UEs, etc.), for example, based on solicited or unsolicited requests, and then report such measurements to the target UE 104a.

[0266] At 1406 , other respective UEs 104 b and / or the network entity 102 may generate a training data set based on measurement reports from different sources and perform training of the AI / ML model.

[0267] At 1408, depending on the implementation, other respective UEs 104b and / or network entities 102 may perform inference based on the new measurement data by repeating the processes at 1402, 1404. In implementations, the training and inference datasets may be requested at a single time, such as as an alternative to repeating 1402, 1404.

[0268] In implementation, without performing AI / ML model training on the UE side, various network entities or nodes may be enabled using the following procedures to enable reporting configuration:

[0269] The target UE performing the training may send a direct AI / ML positioning reporting configuration to a network entity and / or other UEs / devices and receive corresponding reports of downlink (DL) or sidelink (SL) direct AI / ML positioning measurements (e.g., fingerprint measurements). In other implementations, UL positioning measurements may also be provided, such as where the UE may train a model based on such UL measurements (e.g., fingerprints).

[0270] The network entity performing the training (e.g., location server and / or NG-RAN node) may send a direct AI / ML positioning reporting configuration to the target UE and receive a report for downlink (DL) or sidelink (SL) direct AI / ML positioning measurements (e.g.,

[0271] Fingerprint measurement) of the corresponding report.

[0272] ●The UE / node performing training (e.g., anchor UE, PRU UE, SL UE, etc.) can send direct AI / ML positioning report configuration to the target UE and receive corresponding reports for downlink (DL) or sidelink (SL) direct AI / ML positioning measurements (e.g., fingerprint measurements).

[0273] In scenarios for UE-assisted positioning, the location server may request measurement training data sets from various data sources based on DL and SL reference signals (e.g., DL-PRS, SL-PRS, etc.), including measurements collected internally within the location server, multiple UEs, other neighboring location servers, NG-RAN nodes, positioning reference units, and combinations thereof.

[0274] Figure 15 A scenario 1500 is illustrated for supporting machine learning for positioning according to aspects of the present disclosure. For example, scenario 1500 represents an implementation of report exchange when performing network-side training and network-side inference. In scenario 1500, training and inference are performed at a network entity 102 (e.g., a location server), and the network entity 102 can employ the signaling mechanism shown to signal a direct AI / ML positioning report configuration and receive measurements from a UE 104 (e.g., a target UE, a PRU UE, etc.).

[0275] At 1502, the corresponding network entity 102 may request a plurality of direct AI / ML positioning measurements based on certain defined reporting criteria from the UE 104. As for signaling for transmitting the request 1502 to the UE 104, LPP signaling may be employed, while in the scenario for an NG-RAN node, RRC and / or related DL signaling, such as DL MAC CE, may be employed.

[0276] At 1504 , UE 104 may respond in the same manner to request 1502 received from network entity 102 and provide measurements as part of response 1504 according to the reporting criteria and / or assistance information specified by request 1502 .

[0277] At 1506, the network entity 102 constructs a training dataset based on the measurement reports from the different UEs 104 and performs training of the AI / ML model. At 1508, the network entity 102 performs inference based on the new measurement data by repeating steps 1502 and 1504. Alternatively or additionally, the training and inference datasets may be requested in a single request.

[0278] Figure 16 A scenario 1600 is illustrated for supporting machine learning for positioning according to aspects of the present disclosure. For example, scenario 1600 illustrates an implementation of a request and response process for a stored AI / ML training dataset based on reported AI / ML measurements. In scenario 1500 where training is performed at a UE 104 (e.g., a PRU UE), the UE may employ the signaling shown in scenario 1600 and described below to receive the AI / ML training dataset for training at the UE 104.

[0279] At 1602, the network entity 102 may request a plurality of direct AI / ML positioning measurements from the UE 104 based on a defined reporting criterion. As for signaling transmission from the network entity 102 to the UE 104, LPP signaling may be adopted, and in the scenario for an NG-RAN node, RRC and / or related DL signaling, such as DL MAC CE, may be adopted.

[0280] At 1604, the UE 104 may respond to the request 1602 from the network entity and provide measurements according to the reporting criteria and / or assistance information specified in the request 1602. At 1606, the network entity 102 constructs a training dataset based on the measurement reports from the different source UEs / nodes and stores the AI / ML training dataset. The storage of the training dataset may be associated with additional validity criteria, such as time criteria (e.g., time window, validity period, timer expiration) and / or spatial criteria, such as geographic area ID, area ID, cell ID, tracking area ID, system information area ID, segment ID, AI-ML dataset validity area, or a combination thereof.

[0281] At 1608, in a scenario where training is performed on the UE 104 side, the UE 104 may request an AI / ML training dataset that is based in part on measurements provided from 1604. At 1610, the network entity 102 may respond to the request 1608 with the AI / ML training dataset, which may be based on certain criteria including the suitability of the training dataset to the UE 104 location, radio link quality, radio channel parameters, mobility pattern, orientation, etc.

[0282] At 1612 , the UE 104 performs training based on the training data set received from the network entity 102 , and at 1614 , the network entity 102 and / or the UE 104 may perform inference to derive an estimated location of the target UE 104 based on the trained AI / ML model.

[0283] In implementation, scenario 1600 may be applicable to a scenario where, if training is performed at an NG-RAN node, the NG-RAN node may request and receive a constructed training data set stored at a location server (e.g., LMF) based in part on measurements provided by the NG-RAN to the location server, as described below with reference to Figure 17 As stated.

[0284] In implementation, model training can be performed on the UE side, where the signaling mechanism in scenario 1300b can be utilized to signal direct AI / ML positioning reporting configuration and receive measurements from network entities. One or more UEs 104 (e.g., target UEs) can provide multiple direct AI / ML measurements for inference on the network side.

[0285] In implementation, in scenarios where AI / ML model training and inference are performed on the network side, various network entities and / or nodes may be enabled using the following procedures to enable reporting configuration. For example, a network entity (e.g., a location server) may send a direct AI / ML positioning reporting configuration to a UE and receive a corresponding report for DL ​​and / or SL direct AI / ML positioning measurements (e.g., fingerprint measurements).

[0286] In scenarios for NG-RAN assisted positioning, the location server may request multiple UL direct AI / ML positioning measurements based on the UL reference signals used for positioning (e.g., SRS) from various data sources, including neighboring gNB / TRP or NG-RAN nodes, PRU TRP, CU, DU, or a combination thereof.

[0287] Figure 17 A scenario 1700 is illustrated for supporting machine learning for positioning according to aspects of the present disclosure. For example, scenario 1700 illustrates a report exchange when LMF-side training is performed with LMF-side inference. In scenario 1700 where training and inference are performed at a location server 1702, the location server 1702 may employ the signaling mechanism shown to signal UL direct AI / ML positioning reports.

[0288] At 1704, the location server 1702 may request a plurality of UL direct AI / ML positioning measurements from the NG-RAN node 1706 based on defined reporting criteria, such as serving gNB / TRP, neighboring gNB / TCP, PRU gNB, TRP, etc. Signaling to the NG-RAN node 1706 may employ NRPPa signaling, such as Positioning Measurement Request and Positioning Measurement Response messages.

[0289] At 1708 , the NG-RAN node 1706 may respond to the request(s) 1704 from the location server 1702 and provide measurements according to the reporting criteria and / or assistance information, such as specified in the request 1704 .

[0290] At 1710 , the location server 1702 constructs a training data set based on measurement reports from different source NG-RAN nodes 1706 and performs training of the AI / ML model.

[0291] At 1712, the location server 1702 performs inference based on the new measurement data by repeating steps 1704, 1706. In implementations, the training and inference data sets may be requested in a single request.

[0292] In implementations where training is performed at the NG-RAN node 1706, the NG-RAN node 1706 may request and receive UL direct AI / ML positioning measurement reports from the location server 1702, e.g., via NRPPa and / or other NG-RAN nodes (e.g., via an Xn interface).

[0293] In an implementation, various network entities and / or nodes may be enabled using the following process: the location server 1702 may send direct AI / ML positioning reporting configurations to multiple NG-RAN nodes 1706 and receive corresponding reports for UL direct AI / ML positioning measurements (e.g., fingerprint measurements).

[0294] In additional or alternative implementations, the training dataset construction and dataset training do not necessarily occur at the same entity, but may also occur at other separate network entities or nodes. For example, according to scenarios 1400a to 1400c, the PRU UE, anchor UE, and / or other UEs may perform the dataset construction or dataset training. Furthermore, similar to scenario 1600, the serving gNB / TRP, neighboring gNB / TRB, or PRU gNB / TCP may all perform the training dataset construction or dataset training. Furthermore, the training dataset construction and inference of the AI / ML model may also follow the same behavior, so they do not necessarily need to be performed at the same entity. Alternatively or additionally, the above-described reporting configuration methods may be combined in various ways, such as utilizing multiple DL, SL, and UL direct AI / ML positioning measurements in any one or more combinations.

[0295] The implementation described in this paper also provides various ML-related reporting standards. For example, configuration of reporting standards is detailed to support direct AI / ML position estimation, such as using fingerprinting methods.

[0296] In UE-based positioning scenarios (e.g., utilizing UE-side models), the target UE may request multiple DL or SL direct AI / ML positioning measurements, or other data required for training or inference. A set of common reporting standards may be defined for network entities or UEs / devices to provide measurement reports for dataset construction.

[0297] Table 9: Public reporting standards

[0298]

[0299]

[0300]

[0301]

[0302]

[0303] For the UE-assisted positioning scenario, one or more common reporting criteria detailed in Table 9 may be signaled by a network entity (eg, a location server for reporting data types (eg, measurement data)).

[0304] For NG-RAN assisted positioning scenarios, one or more common reporting criteria detailed in Table 9 may be signaled by a network entity (e.g., location server) to an NG-RAN node (e.g., gNB) for reporting data types, such as measurement data.

[0305] In an implementation, low-layer signaling (e.g., RRC signaling, MAC CE, DCI signaling, or a combination thereof) regarding LPP may be used to convey the direct AI / ML reporting standard configuration. In an implementation, LPP or RRC signaling may be used to add, modify, remove, update, activate, and / or deactivate the direct AI / ML positioning reporting configuration for one or more UEs.

[0306] In implementation, the reporting criteria and associated implementation details indicated in Table 9 may be extended to assisted AI / ML positioning measurements for cases A, B, and C, where one or more AI / ML models are used to enhance positioning measurements.

[0307] In implementation, common reporting criteria may be broadcast to multiple UEs within a given geographic area (e.g., based on the same cell ID, based on system information area, based on sector ID, or a combination thereof) via system information block (SIB) or positioning system information block (posSIB) messages.

[0308] The implementation also provides for reporting various auxiliary information related to direct AI / ML positioning measurements to help derive the position estimate of the target UE. For example, this can be extended to report scenarios where AI / ML assisted positioning measurements are also required (if applicable).

[0309] In an implementation, a measurement entity (e.g., a UE or NG-RAN node performing AI / ML positioning measurements) may be configured to report positioning measurement correlation between different sets of measurements performed at the same measurement entity. For example, the correlation metric may be affected by UE capabilities. In at least one implementation, the measurement entity may report RSS correlation associated with a set of AI / ML positioning RSRP / RSI measurements for each DL or SL PRS resource (e.g., DL resource ID) via higher layer parameters (e.g., LPP signaling). In scenarios for NG-RAN node measurements, RSS correlation may be associated with a set of different UL RSS measurements for each UL resource (e.g., SRS resource ID). In an extended implementation, the correlation metric may be applicable to timing-based (e.g., RSTD, ToA, etc.) or angle-based (e.g., AoA, AoD, etc.) measurements. The correlation measurement metric may be obtained for each ground-truth reference location to accurately and fairly calculate measurement correlation for multiple measurements performed at the same ground-truth reference location.

[0310] In an implementation, positioning measurement correlations may be obtained from different UEs / devices at the same ground truth reference location. The measurements may be based at least in part on UE / NG-RAN node provider-specific variations, and thus variations may exist for the same positioning measurements at the same ground truth reference location. In such an implementation, the network entity or UE / device collecting the positioning measurements may correlate the different measurements received from different network nodes / UEs / devices. In addition to the spatial domain, correlation may also be performed in the time domain (e.g., using a (sliding) time window), e.g., based on location. In an implementation, measurement correlations between adjacent reference location points may also be configured, determined, and reported.

[0311] Depending on the implementation, the measurement entity (e.g., UE and / or NG-RAN node) performing the AI / ML positioning measurement may be configured to report the associated channel characteristics of the direct AI / ML positioning measurement, including whether the measurement is line-of-sight (LOS) or non-line-of-sight (NLOS), based on binary (e.g., hard decision) or soft indicators, link path loss, channel coefficients, or a combination thereof. Reporting additional channel characteristics associated with the measurement may improve the stability and reliability of the reported AI / ML positioning measurements (e.g., RSS measurements such as RSRP).

[0312] In an implementation, the difference in path loss between the ground truth reference location point direct AI / ML (e.g., fingerprint measurement) and the target UE's measurement can be used to derive PRS / SRS RSS measurements based on Tx and Rx antenna gains, path loss reference, path loss exponent, standard deviation of fading parameters (e.g., shadow fading at (multiple) ground truth reference location points and at the unknown target UE location). One or more of the above parameters can be configured for reporting and reported to a requesting entity, such as a UE / device or NG-RAN node or location server. In an implementation, the measuring entity and / or the target UE can calculate the path loss and report it to the requesting entity along with the direct AI / ML positioning measurement.

[0313] In implementation, the measurement entity (e.g., UE or NG-RAN node) performing AI / ML positioning measurements may be configured to report (N×M i ) j The average measurement at each ground truth reference location point over N sampling points, where N is the number of configured samples per measurement instance and M is the total number of measurements from each igNB / TRP in case of DL positioning measurements at each j-th ground truth reference location, and M is the total number of measurements collected from each i-th UE in case of UL measurements. In implementation, this can be done over N×M i Additional statistical measurements are obtained at each reference position / point, including variance, standard deviation, probability distribution function, cumulative distribution function, etc.i It can also be configured using higher layer signaling (such as LPP, RRC, SLPP or a combination thereof).

[0314] In implementation, when the configured direct AI / ML positioning model includes k-NN in the case of supervision or K-means in the case of an unsupervised model, the network entity that uses inference to determine the location of the target UE based on the fingerprint dataset may use the following generalized distance formula based on the Minkowski distance to derive the location of the target UE by processing the newly received measurements using the following formula:

[0315]

[0316] Where n is the total number of received measurements with parameter pairs (x, y), and a can be configured according to the distance algorithm utilized, for example, if a=1, the Manhattan distance method is used, while if a=2, the Euclidean distance method is used. In other implementations, Hamming distance or cosine distance and cosine similarity can be utilized to determine the similarity between multi-dimensional direct AI / ML positioning data.

[0317] In implementation, the online measurements will be matched with the fingerprint measurements at each ground truth reference location / point, and a similarity score based on the cumulative Manhattan distance in equation (1) where a=1 can be utilized to determine the location of the target UE, whereby the measurement entity is configured to report a total of M i，j The minimum and maximum PRS / SRS RSS measurements among the measurements, where i refers to the measurement initiated from each i-th gNB / TRP or UE at each j-th ground truth reference location. The similarity score (β Ref-Location ) can be given by the following mathematical relationship, where:

[0318]

[0319] in is the minimum PRS or SRS RSS / fingerprint positioning measurement from the igNB / TRP or UE, and is the minimum sample positioning measurement in the set of measurements provided by the target UE, is the maximum PRS or SRS RSS / fingerprint positioning measurement from the igNB / TRP or UE, and is the maximum sample positioning measurement in the set of measurements provided by the target UE. Ref-Location The minimum value of (j) corresponds to the most likely position where the target UE may be located.

[0320] In the implementation, β Ref-Location(j) can be derived from a predefined time window associated with a start time, window length, end time, periodicity to capture the variations in positioning measurements and thereby the similarity score over time.

[0321] In implementation, the first arrival path is considered for the above RSS measurement to be used as part of the fingerprint recognition training dataset. In implementation, the first arrival path and up to T configurable additional paths may be associated with the fingerprint RSS measurement and may be reported to the requesting network entity / node / UE.

[0322] In an implementation, a measurement entity (e.g., UE, PRU UE, or SL UE) may be configured to indicate whether RSS / fingerprint measurements (e.g., DL PRS RSRP) from a set of configured PRS resources within the same resource set have been measured using the same DL receive beam or the same spatial filter.

[0323] In implementations, the measurement, training, and / or inference entities may be configured to self-calibrate direct or assisted AI / ML positioning measurements based on the provision of certain parameters for inclusion in a training or inference dataset for a reference device (e.g., a PRU UE). In at least one example, a linear calibration may be employed to align measurements of a target UE with measurements of a reference device (such as a PRU UE), which may be expressed as follows:

[0324]

[0325] in represents the average target UE PRS / SRS positioning measurement from the i-th gNB / TRP or UE at each j-th ground truth reference position, represents the average PRU UE PRS / SRS positioning measurement from the ig NB / TRP or UE at each j-th ground truth reference position, while δ TP and μ TP is a linear calibration parameter used to map RSS measurements from the target UE to the PRU UE. This is particularly useful if different network entities or UEs / devices are performing measurements from different providers. The linear parameter δ TP and μ TP The measurement entity may be configured via higher layer signaling (e.g., LPP, NRPPa, SLPP, etc.) to calibrate the measurement before reporting. The measurement, training, and / or inference entity may also receive a request to perform self-calibration of direct or assisted AI / ML positioning measurements. In another implementation, a nonlinear function may also be utilized to self-calibrate direct or assisted AI / ML positioning measurements, utilizing a similar process outlined for linear self-calibration in terms of supplying and reporting nonlinear self-calibration parameters.

[0326] The RSS measurements mentioned in the implementations described herein may include RSRP, RSRPP, RSSI, RSRQ values ​​associated with DL PRS, SL PRS, or UL SRS.

[0327] Figure 18 An example of a block diagram 1800 of a device 1802 (e.g., an apparatus) supporting machine learning for positioning according to aspects of the present disclosure is illustrated. The device 1802 can be an example of a UE 104 as described herein. The device 1802 can support wireless communications with one or more network entities 102, UEs 104, or any combination thereof. The device 1802 can include components for two-way communication, including components for sending and receiving communications (such as a processor 1804, a memory 1806, a transceiver 1808, and an I / O controller 1810). These components can be in electronic communication or otherwise coupled (e.g., operatively, communicatively, functionally, electronically, electrically) via one or more interfaces (e.g., a bus).

[0328] The processor 1804, memory 1806, transceiver 1808, or various combinations thereof, or various components thereof, may be examples of means for performing various aspects of the present disclosure described herein. For example, the processor 1804, memory 1806, transceiver 1808, or various combinations thereof, or components thereof, may support a method for performing one or more of the operations described herein.

[0329] In some implementations, the processor 1804, memory 1806, transceiver 1808, or various combinations or components thereof may be implemented in hardware (e.g., in a communications management circuit system). The hardware may include a processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination of components configured to or otherwise support the functions described in this disclosure. In some implementations, the processor 1804 and the memory 1806 coupled to the processor 1804 may be configured to perform one or more functions described herein (e.g., execution of instructions stored in the memory 1806 by the processor 1804). For example, in the context of a UE 104, the transceiver 1808 and the processor 1804 coupled to the transceiver 1808 are configured to cause the UE 104 to perform the various described operations and / or combinations thereof.

[0330] For example, according to examples disclosed herein, the processor 1804 and / or the transceiver 1808 may support wireless communications at the device 1802. For example, the processor 1804 and / or the transceiver 1808 may be configured to and / or otherwise support components for: receiving a machine learning positioning configuration request; and sending a machine learning positioning configuration response based at least in part on the machine learning positioning configuration request, the machine learning positioning configuration response including a positioning reference signal configuration to be measured at a corresponding reference location and utilizing one or more associated validity criteria.

[0331] In addition, in some implementations, the processor is configured to cause the device to: receive a machine learning positioning configuration request from one or more of a network node or a user equipment (UE) and send a machine learning positioning configuration response to one or more of the network node or the UE; the reference location includes one or more ground truth reference locations; the machine learning positioning configuration response includes an artificial intelligence configuration for a positioning reference signal configuration; the processor is configured to cause the device to: send one or more machine learning positioning configuration responses independently of the machine learning positioning configuration request; the machine learning positioning configuration response includes one or more of the following items: a direct machine learning configuration or an assisted machine learning configuration; the direct machine learning configuration includes a configuration for performing radio frequency fingerprinting.

[0332] In addition, in some implementations, the machine learning positioning configuration request includes a request for one or more of the following: a training type, a ground truth position request, a machine learning method, a machine learning on-demand positioning reference signal request, or pre-configured machine learning assistance data; the positioning reference signal configuration includes one or more of the following: positioning reference signal resources, resource sets, transmission reception points, or positioning frequency layers to be measured at each reference location; one or more associated validity criteria include one or more of the following: a time standard or a spatial standard; the time standard includes a measurement time defined by one or more time bases; the one or more time bases include at least one of the following: a system frame number, coordinated universal time (UTC), or global navigation satellite system (GNSS) time; one or more associated validity criteria include a spatial standard relating to an indication of a geographic area, and the spatial standard includes one or more of the following: an area identifier, a cell identifier, or a segment identifier.

[0333] In addition, in some implementations, the positioning reference signal configuration includes one or more indications of one or more of the following items: uplink radio access technology-related measurements, downlink radio access technology-related measurements, or sidelink radio access technology-related measurements to be performed; the positioning reference signal configuration includes one or more indications of one or more of the following items: downlink radio access technology-independent measurements or sidelink radio access technology-independent measurements to be performed; the machine learning positioning configuration response also includes one or more configurations related to at least one of the following items: k-nearest neighbor (k-NN) algorithm, support vector machine, decision tree, random forest, Gaussian mixture model, Bayesian network, artificial neural network, k-means, reinforcement learning, deep learning, or transfer learning; the device includes at least one of the following items: a location server, a next generation radio access network (NG-RAN), a positioning reference unit (PRU), a user equipment (UE), an anchor UE, or a target UE.

[0334] In addition, in some implementations, the positioning reference signal configuration includes a received signal strength indicator measurement for one or more of the downlink, uplink, or sidelink; and the received signal strength indicator measurement for one or more of the downlink, uplink, or sidelink is defined as: comprising a linear average of the total received power observed in the resource elements of the time slot carrying the positioning reference signal configured for measurement; the positioning reference signal configuration includes a positioning reference signal arrival time measurement for one or more of the downlink, uplink, or sidelink; and the positioning reference signal arrival time measurement for one or more of the downlink, uplink, or sidelink is defined as comprising the reception time of the positioning reference signal at the receiver reference point; the positioning reference signal configuration includes an indication that the location server is permitted to send an error cause related to an incorrect configuration of the positioning reference signal configuration; the positioning reference signal configuration includes an indication that the target user equipment (UE) is permitted to send an error cause related to one or more of the following items: an error in receiving a machine learning positioning configuration response, or a measurement error related to the machine learning positioning measurement.

[0335] In another example, according to examples disclosed herein, the processor 1804 and / or the transceiver 1808 can support wireless communications at the device 1802. For example, the processor 1804 and / or the transceiver 1808 can be configured to or otherwise support components for: sending a machine learning position configuration request; receiving a machine learning position configuration response including a positioning reference signal configuration to be measured at one or more respective reference locations and utilizing one or more associated validity criteria; and performing one or more machine learning position measurements based at least in part on the positioning reference signal configuration.

[0336] In addition, in some implementations, the processor and transceiver are configured to cause the apparatus to do one or more of: receive a machine learning positioning configuration response in response to a machine learning positioning configuration request; or receive a machine learning positioning configuration response independent of a machine learning positioning configuration request; the processor is configured to cause the apparatus to: input one or more machine learning position measurements into a machine learning model and receive output from the machine learning model; the processor is configured to cause the apparatus to generate an estimated position of the apparatus based at least in part on the output from the machine learning model; the apparatus includes one or more of: a target user equipment (UE), a positioning reference unit (PRU), or an anchor UE.

[0337] In another example, according to examples disclosed herein, the processor 1804 and / or the transceiver 1808 can support wireless communications at the device 1802. For example, the processor 1804 and / or the transceiver 1808 can be configured to or otherwise support components for: sending a configuration request to configure a reference signal for machine learning positioning measurements; receiving a configuration response including the reference signal configuration; and sending a reference signal transmission activation command based at least in part on the reference signal configuration.

[0338] In addition, in some implementations, the reference signal includes one or more of the following items: a sounding reference signal or a positioning reference signal; the processor is configured to cause the device to send a reference signal transmission deactivation command; the device includes a location server, and wherein the processor is configured to cause the device to send a reference signal transmission activation command to one or more other devices configured to send a reference signal.

[0339] In further examples, according to examples disclosed herein, the processor 1804 and / or the transceiver 1808 can support wireless communications at the device 1802. For example, the processor 1804 and / or the transceiver 1808 can be configured to or otherwise support components for: sending one or more machine-learned positioning report requests, the machine-learned positioning report requests including a machine-learned reporting configuration and one or more common reporting criteria; receiving one or more machine-learned positioning reports; processing the one or more machine-learned positioning reports via a machine-learned model; and generating an estimated position of a user equipment (UE) based at least in part on an output from the machine-learned model.

[0340] In addition, in some implementations, the machine learning report configuration includes one or more of the following items: direct machine learning report configuration or assisted machine learning report configuration; the device includes a configuration entity, and wherein the configuration entity includes at least one of the following items: a location server, a next generation radio access network (NG-RAN), a positioning reference unit (PRU), a UE, an anchor UE, or a target UE; the processor is configured to cause the device to send one or more machine learning positioning report requests to one or more second devices, and receive one or more machine learning positioning reports from one or more second devices, and wherein the one or more second devices include at least one of the following items: a location server, a next generation radio access network (NG-RAN), a positioning reference unit (PRU), a UE, an anchor UE, or a target UE.

[0341] In addition, in some implementations, one or more public reporting standards include one or more of the following items: fingerprint type, ground truth reference position type, time domain reporting type, measurement preprocessing, measurement validity, UE type, mobility indication, orientation indication, fingerprint recognition environment indication, fingerprint quality indication or tag quality indication; for at least one machine learning positioning report request, one or more public reporting standards in the public reporting standards are configured to be added, removed, updated, activated or deactivated; the processor is configured to cause the device to broadcast the one or more public reporting standards via positioning system information broadcast signaling; the machine learning report configuration includes an indication of reporting machine learning positioning measurement correlation between different measurement sets.

[0342] In addition, in some implementations, the machine learning positioning measurement correlation includes one or more of the following: spatial correlation or time domain correlation; one or more machine learning positioning reports include one or more of the following: machine learning positioning measurements or machine learning positioning location information; the machine learning report configuration includes an indication to report path loss at different locations, the different locations including a ground truth reference location; the machine learning report configuration includes an indication to average the machine learning positioning measurements over a configured number of measurements and report the average as part of one or more machine learning positioning reports; the machine learning report configuration includes an indication to determine a similarity score at configured locations to determine an optimal mapping between the fingerprint measurement and the estimated location of the target UE.

[0343] In further examples, according to examples disclosed herein, the processor 1804 and / or the transceiver 1808 can support wireless communications at the device 1802. For example, the processor 1804 and / or the transceiver 1808 can be configured to or otherwise support components for: receiving one or more machine learning position report requests, the machine learning position report requests including a machine learning report configuration and one or more public reporting standards; generating one or more machine learning position reports based at least in part on the machine learning report configuration and the one or more public reporting standards; and sending the one or more machine learning position reports.

[0344] In addition, in some implementations, the apparatus includes at least one of the following: a user equipment (UE), an anchor UE, or a target UE; the apparatus includes a configuration entity, and wherein the configuration entity includes at least one of the following: a location server, a next generation radio access network (NG-RAN), or a positioning reference unit (PRU).

[0345] In further examples, according to examples disclosed herein, the processor 1804 and / or the transceiver 1808 can support wireless communications at the device 1802. For example, the processor 1804 and / or the transceiver 1808 can be configured to or otherwise support components for: sending one or more machine learning location report requests, the machine learning location report requests including a machine learning report configuration and one or more common reporting criteria; receiving one or more machine learning location reports; generating a machine learning location training dataset based at least in part on the one or more machine learning location reports; and training a location machine learning model using the machine learning location training dataset to generate a trained location machine learning model.

[0346] In addition, in some implementations, the processor and transceiver are configured to cause the device to: receive one or more additional machine learning positioning reports; input at least a portion of the one or more additional machine learning positioning reports into a trained positioning machine learning model; and estimate the location of a target user equipment (UE) based at least in part on the output from the trained positioning machine learning model; the processor is configured to cause the device to: receive a request for machine learning positioning training data for positioning; and based at least in part on the request, send a machine learning positioning training data set; the processor is configured to cause the device to: receive a request for a trained positioning machine learning model; and based at least in part on the request, send at least a portion of the trained positioning machine learning model.

[0347] In another example, according to examples disclosed herein, the processor 1804 and / or the transceiver 1808 can support wireless communications at the device 1802. For example, the processor 1804 and / or the transceiver 1808 can be configured to or otherwise support components for: receiving a machine learning positioning configuration request; and sending a machine learning positioning configuration response based at least in part on the machine learning positioning configuration request, the machine learning positioning configuration response including a positioning reference signal configuration to be measured at a corresponding reference location and utilizing one or more associated validity criteria.

[0348] In addition, in some implementations, for example, the processor 1804 and / or the transceiver 1808 may be configured to or otherwise support components for: receiving a machine learning positioning configuration request from one or more network nodes or user equipment (UEs) and sending a machine learning positioning configuration response to one or more network nodes or UEs; the reference location includes one or more ground truth reference locations; the machine learning positioning configuration response includes an artificial intelligence configuration for a positioning reference signal configuration; one or more machine learning positioning configuration responses are sent independently of the machine learning positioning configuration request; the machine learning positioning configuration response includes one or more of the following: a direct machine learning configuration or an assisted machine learning configuration; the direct machine learning configuration includes a configuration for performing radio frequency fingerprinting.

[0349] In addition, in some implementations, the machine learning positioning configuration request includes a request for one or more of the following: a training type, a ground truth position request, a machine learning method, a machine learning on-demand positioning reference signal request, or pre-configured machine learning assistance data; the positioning reference signal configuration includes one or more of the following: positioning reference signal resources, resource sets, transmission reception points, or positioning frequency layers to be measured at each reference location; one or more associated validity criteria include one or more of the following: a time standard or a spatial standard; the time standard includes a measurement time defined by one or more time bases; the one or more time bases include at least one of the following: a system frame number, coordinated universal time (UTC), or global navigation satellite system (GNSS) time; the one or more associated validity criteria include a spatial standard relating to an indication of a geographic area, and the spatial standard includes one or more of the following: an area identifier, a cell identifier, or a segment identifier.

[0350] In addition, in some implementations, the positioning reference signal configuration includes one or more indications of one or more of the following items: uplink radio access technology-related measurements, downlink radio access technology-related measurements, or sidelink radio access technology-related measurements to be performed; the positioning reference signal configuration includes one or more indications of one or more of the following items: downlink radio access technology-independent measurements or sidelink radio access technology-independent measurements to be performed; the machine learning positioning configuration response also includes one or more configurations related to at least one of the following items: k-nearest neighbor (k-NN) algorithm, support vector machine, decision tree, random forest, Gaussian mixture model, Bayesian network, artificial neural network, k-means, reinforcement learning, deep learning, or transfer learning; the method is performed by an apparatus comprising at least one of the following items: a location server, a next generation radio access network (NG-RAN), a positioning reference unit (PRU), a user equipment (UE), an anchor UE, or a target UE.

[0351] In addition, in some implementations, the positioning reference signal configuration includes a received signal strength indicator measurement for one or more of the downlink, uplink, or sidelink; and the received signal strength indicator measurement for one or more of the downlink, uplink, or sidelink is defined as: comprising a linear average of the total received power observed in the resource elements of the time slot carrying the positioning reference signal configured for measurement; the positioning reference signal configuration includes a positioning reference signal arrival time measurement for one or more of the downlink, uplink, or sidelink; and the positioning reference signal arrival time measurement for one or more of the downlink, uplink, or sidelink is defined as comprising the reception time of the positioning reference signal at the receiver reference point; the positioning reference signal configuration includes an indication that the location server is permitted to send an error cause related to an incorrect configuration of the positioning reference signal configuration; the positioning reference signal configuration includes an indication that the target user equipment (UE) is permitted to send an error cause related to one or more of the following items: an error in receiving a machine learning positioning configuration response, or a measurement error related to the machine learning positioning measurement.

[0352] In further examples, according to examples disclosed herein, the processor 1804 and / or the transceiver 1808 can support wireless communications at the device 1802. For example, the processor 1804 and / or the transceiver 1808 can be configured to or otherwise support components for: sending a machine learning position configuration request; receiving a machine learning position configuration response including a positioning reference signal configuration to be measured at one or more respective reference locations and utilizing one or more associated validity criteria; and performing one or more machine learning position measurements based at least in part on the positioning reference signal configuration.

[0353] In addition, in some implementations, for example, the processor 1804 and / or the transceiver 1808 may be configured to or otherwise support components for one or more of: receiving a machine learning positioning configuration response in response to a machine learning positioning configuration request; or receiving a machine learning positioning configuration response independent of the machine learning positioning configuration request; inputting one or more machine learning position measurements into a machine learning model and receiving output from the machine learning model; generating an estimated position of the device based at least in part on the output from the machine learning model; the method being performed by a device comprising one or more of: a target user equipment (UE), a positioning reference unit (PRU), or an anchor UE.

[0354] In another example, according to examples disclosed herein, the processor 1804 and / or the transceiver 1808 can support wireless communications at the device 1802. For example, the processor 1804 and / or the transceiver 1808 can be configured to or otherwise support components for: sending a configuration request to configure a reference signal for machine learning positioning measurements; receiving a configuration response including the reference signal configuration; and sending a reference signal transmission activation command based at least in part on the reference signal configuration.

[0355] In addition, in some implementations, the reference signal includes one or more of the following items: a sounding reference signal or a positioning reference signal; sending a reference signal transmission deactivation command; the method is performed by an apparatus, the apparatus including a location server, and wherein the method also includes sending a reference signal transmission activation command to one or more other apparatuses configured to send a reference signal.

[0356] In further examples, according to examples disclosed herein, the processor 1804 and / or the transceiver 1808 can support wireless communications at the device 1802. For example, the processor 1804 and / or the transceiver 1808 can be configured to or otherwise support components for: sending one or more machine-learned positioning report requests, the machine-learned positioning report requests including a machine-learned reporting configuration and one or more common reporting criteria; receiving one or more machine-learned positioning reports; processing the one or more machine-learned positioning reports via a machine-learned model; and generating an estimated position of a user equipment (UE) based at least in part on an output from the machine-learned model.

[0357] In addition, in some implementations, the machine learning report configuration includes one or more of the following items: direct machine learning report configuration or assisted machine learning report configuration; wherein the method is performed by an apparatus, the apparatus including a configuration entity, and wherein the configuration entity includes at least one of the following items: a location server, a next generation radio access network (NG-RAN), a positioning reference unit (PRU), a UE, an anchor UE, or a target UE; the method also includes: sending one or more machine learning positioning report requests from the first apparatus to one or more second apparatuses, and receiving one or more machine learning positioning reports from the one or more second apparatuses, and wherein the one or more second apparatuses include at least one of the following items: a location server, a next generation radio access network (NG-RAN), a positioning reference unit (PRU), a UE, an anchor UE, or a target UE.

[0358] In addition, in some implementations, one or more public reporting standards include one or more of the following items: fingerprint type, ground truth reference position type, time domain reporting type, measurement preprocessing, measurement validity, UE type, mobility indication, orientation indication, fingerprint recognition environment indication, fingerprint quality indication or tag quality indication; wherein for at least one machine learning positioning report request, one or more public reporting standards in the public reporting standards are configured to be added, removed, updated, activated or deactivated; the method also includes broadcasting one or more public reporting standards via positioning system information broadcast signaling; wherein the machine learning report configuration includes reporting an indication of the correlation of machine learning positioning measurements between different measurement sets.

[0359] In addition, in some implementations, the machine learning positioning measurement correlation includes one or more of the following: spatial correlation or time domain correlation; wherein the one or more machine learning positioning reports include one or more of the following: machine learning positioning measurements or machine learning positioning location information; wherein the machine learning report configuration includes an indication of reporting path loss at different locations, the different locations including a ground truth reference location; wherein the machine learning report configuration includes an indication of averaging the machine learning positioning measurements over a configured number of measurements and reporting the average as part of the one or more machine learning positioning reports; wherein the machine learning report configuration includes an indication of determining a similarity score at configured locations to determine an optimal mapping between the fingerprint measurement and the estimated location of the target UE.

[0360] In further examples, according to examples disclosed herein, the processor 1804 and / or the transceiver 1808 can support wireless communications at the device 1802. For example, the processor 1804 and / or the transceiver 1808 can be configured to or otherwise support components for: receiving one or more machine learning position report requests, the machine learning position report requests including a machine learning report configuration and one or more public reporting standards; generating one or more machine learning position reports based at least in part on the machine learning report configuration and the one or more public reporting standards; and sending the one or more machine learning position reports.

[0361] In addition, in some implementations, the method is performed by an apparatus comprising one or more of: a user equipment (UE), an anchor UE, or a target UE; wherein the method is performed by an apparatus comprising a configuration entity, and wherein the configuration entity comprises at least one of: a location server, a next generation radio access network (NG-RAN), or a positioning reference unit (PRU).

[0362] In further examples, according to examples disclosed herein, the processor 1804 and / or the transceiver 1808 can support wireless communications at the device 1802. For example, the processor 1804 and / or the transceiver 1808 can be configured to or otherwise support components for: sending one or more machine learning location report requests, the machine learning location report requests including a machine learning report configuration and one or more common reporting criteria; receiving one or more machine learning location reports; generating a machine learning location training dataset based at least in part on the one or more machine learning location reports; and training a location machine learning model using the machine learning location training dataset to generate a trained location machine learning model.

[0363] In addition, in some implementations, for example, the processor 1804 and / or the transceiver 1808 can be configured to or otherwise support components for: receiving one or more additional machine learning positioning reports; inputting at least a portion of the one or more additional machine learning positioning reports into a trained positioning machine learning model; and estimating the location of a target user equipment (UE) based at least in part on an output from the trained positioning machine learning model; the method also includes: receiving a request for machine learning positioning training data for positioning; and sending a machine learning positioning training data set based at least in part on the request; the method also includes: receiving a request for a trained positioning machine learning model; and sending at least a portion of the trained positioning machine learning model based at least in part on the request.

[0364] The processor 1804 may include an intelligent hardware device (e.g., a general-purpose processor, a DSP, a CPU, a microcontroller, an ASIC, an FPGA, a programmable logic device, discrete gate or transistor logic components, discrete hardware components, or any combination thereof). In some implementations, the processor 1804 may be configured to operate a memory array using a memory controller. In some other implementations, the memory controller may be integrated into the processor 1804. The processor 1804 may be configured to execute computer-readable instructions stored in a memory (e.g., memory 1806) to cause the device 1802 to perform various functions of the present disclosure.

[0365] The memory 1806 may include random access memory (RAM) and read-only memory (ROM). The memory 1806 may store computer-readable, computer-executable code, which includes instructions that, when executed by the processor 1804, cause the device 1802 to perform the various functions described herein. The code may be stored in a non-transitory computer-readable medium such as system memory or another type of memory. In some implementations, the code may not be directly executed by the processor 1804, but may cause a computer (e.g., when compiled and executed) to perform the functions described herein. In addition, in some implementations, the memory 1806 may include a basic I / O system (BIOS), which may control basic hardware or software operations, such as interaction with peripheral components or devices.

[0366] I / O controller 1810 can manage input and output signals for device 1802. I / O controller 1810 can also manage peripheral devices that are not integrated into device 2402. In some implementations, I / O controller 1810 can represent a physical connection or port to an external peripheral device. In some implementations, I / O controller 1810 can utilize an operating system, such as MS or other known operating systems. In some implementations, I / O controller 1810 can be implemented as part of a processor such as processor M08. In some implementations, a user can interact with device 1802 via I / O controller 1810 or via hardware components controlled by I / O controller 1810.

[0367] In some implementations, the device 1802 may include a single antenna 1812. However, in some other implementations, the device 1802 may have more than one antenna 1812 (e.g., multiple antennas), including multiple antenna panels or antenna arrays, which may be capable of concurrently sending or receiving multiple wireless transmissions. The transceiver 1808 may communicate bidirectionally via one or more antennas 1812, wired, or wireless links, as described herein. For example, the transceiver 1808 may represent a wireless transceiver and may communicate bidirectionally with another wireless transceiver. The transceiver 1808 may also include a modem that modulates packets to provide the modulated packets to one or more antennas 1812 for transmission, and demodulates packets received from the one or more antennas 1812.

[0368] Figure 19 An example of a block diagram 1900 of a device 1902 (e.g., an apparatus) supporting machine learning for positioning according to aspects of the present disclosure is illustrated. The device 1902 can be an example of a network entity 102 as described herein. The device 1902 can support wireless communication with one or more network entities 102, UEs 104, or any combination thereof. The device 1902 can include components for two-way communication, including components for sending and receiving communications (such as a processor 1904, a memory 1906, a transceiver 1908, and an I / O controller 1910). These components can be in electronic communication or otherwise coupled (e.g., operatively, communicatively, functionally, electronically, electrically) via one or more interfaces (e.g., a bus).

[0369] The processor 1904, memory 1906, transceiver 1908, or various combinations thereof, or various components thereof, may be examples of means for performing various aspects of the present disclosure described herein. For example, the processor 1904, memory 1906, transceiver 1908, or various combinations thereof, or components thereof, may support a method for performing one or more of the operations described herein.

[0370] In some implementations, the processor 1904, memory 1906, transceiver 1908, or various combinations or components thereof may be implemented in hardware (e.g., in a communications management circuit system). The hardware may include a processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination of components configured to or otherwise support the functions described in this disclosure. In some implementations, the processor 1904 and the memory 1906 coupled to the processor 1904 may be configured to perform one or more functions described herein (e.g., the processor 1904 executes instructions stored in the memory 1906). For example, in the context of the network entity 102, the transceiver 1908 and the processor 1904 coupled to the transceiver 1908 are configured to cause the network entity 102 to perform the various described operations and / or combinations thereof.

[0371] For example, according to examples disclosed herein, the processor 1904 and / or the transceiver 1908 may support wireless communications at the device 1902. For example, the processor 1904 and / or the transceiver 1908 may be configured to and / or otherwise support components for: receiving a machine learning positioning configuration request; and sending a machine learning positioning configuration response based at least in part on the machine learning positioning configuration request, the machine learning positioning configuration response including a positioning reference signal configuration to be measured at a corresponding reference location and utilizing one or more associated validity criteria.

[0372] In addition, in some implementations, the processor is configured to cause the device to: receive a machine learning positioning configuration request from one or more of a network node or a user equipment (UE) and send a machine learning positioning configuration response to one or more of the network node or the UE; the reference location includes one or more ground truth reference locations; the machine learning positioning configuration response includes an artificial intelligence configuration for a positioning reference signal configuration; the processor is configured to cause the device to: send one or more machine learning positioning configuration responses independently of the machine learning positioning configuration request; the machine learning positioning configuration response includes one or more of the following items: a direct machine learning configuration or an assisted machine learning configuration; the direct machine learning configuration includes a configuration for performing radio frequency fingerprinting.

[0373] In addition, in some implementations, the machine learning positioning configuration request includes a request for one or more of the following: a training type, a ground truth position request, a machine learning method, a machine learning on-demand positioning reference signal request, or pre-configured machine learning assistance data; the positioning reference signal configuration includes one or more of the following: positioning reference signal resources, resource sets, transmission reception points, or positioning frequency layers to be measured at each reference location; one or more associated validity criteria include one or more of the following: a time standard or a spatial standard; the time standard includes a measurement time defined by one or more time bases; the one or more time bases include at least one of the following: a system frame number, coordinated universal time (UTC), or global navigation satellite system (GNSS) time; one or more associated validity criteria include a spatial standard relating to an indication of a geographic area, and the spatial standard includes one or more of the following: an area identifier, a cell identifier, or a segment identifier.

[0374] In addition, in some implementations, the positioning reference signal configuration includes one or more indications of one or more of the following items: uplink radio access technology-related measurements, downlink radio access technology-related measurements, or sidelink radio access technology-related measurements to be performed; the positioning reference signal configuration includes one or more indications of one or more of the following items: downlink radio access technology-independent measurements or sidelink radio access technology-independent measurements to be performed; the machine learning positioning configuration response also includes one or more configurations related to at least one of the following items: k-nearest neighbor (k-NN) algorithm, support vector machine, decision tree, random forest, Gaussian mixture model, Bayesian network, artificial neural network, k-means, reinforcement learning, deep learning, or transfer learning; the device includes at least one of the following items: a location server, a next generation radio access network (NG-RAN), a positioning reference unit (PRU), a user equipment (UE), an anchor UE, or a target UE.

[0375] In addition, in some implementations, the positioning reference signal configuration includes a received signal strength indicator measurement for one or more of the downlink, uplink, or sidelink; and the received signal strength indicator measurement for one or more of the downlink, uplink, or sidelink is defined as: comprising a linear average of the total received power observed in the resource elements of the time slot carrying the positioning reference signal configured for measurement; the positioning reference signal configuration includes a positioning reference signal arrival time measurement for one or more of the downlink, uplink, or sidelink; and the positioning reference signal arrival time measurement for one or more of the downlink, uplink, or sidelink is defined as comprising the reception time of the positioning reference signal at the receiver reference point; the positioning reference signal configuration includes an indication that the location server is permitted to send an error cause related to an incorrect configuration of the positioning reference signal configuration; the positioning reference signal configuration includes an indication that the target user equipment (UE) is permitted to send an error cause related to one or more of the following items: an error in receiving a machine learning positioning configuration response or a measurement error related to the machine learning positioning measurement.

[0376] In further examples, according to examples disclosed herein, the processor 1904 and / or the transceiver 1908 can support wireless communications at the device 1902. For example, the processor 1904 and / or the transceiver 1908 can be configured to or otherwise support components for: sending a machine learning position configuration request; receiving a machine learning position configuration response including a positioning reference signal configuration to be measured at one or more respective reference locations and utilizing one or more associated validity criteria; and performing one or more machine learning position measurements based at least in part on the positioning reference signal configuration.

[0377] In addition, in some implementations, the processor and transceiver are configured to cause the apparatus to do one or more of: receive a machine learning positioning configuration response in response to a machine learning positioning configuration request; or receive a machine learning positioning configuration response independent of a machine learning positioning configuration request; the processor is configured to cause the apparatus to: input one or more machine learning position measurements into a machine learning model and receive output from the machine learning model; the processor is configured to cause the apparatus to generate an estimated position of the apparatus based at least in part on the output from the machine learning model; the apparatus includes one or more of: a target user equipment (UE), a positioning reference unit (PRU), or an anchor UE.

[0378] In another example, according to examples disclosed herein, the processor 1904 and / or the transceiver 1908 can support wireless communications at the device 1902. For example, the processor 1904 and / or the transceiver 1908 can be configured to or otherwise support components for: sending a configuration request to configure a reference signal for machine learning positioning measurements; receiving a configuration response including a reference signal configuration; and sending a reference signal transmission activation command based at least in part on the reference signal configuration.

[0379] In addition, in some implementations, the reference signal includes one or more of the following items: a sounding reference signal or a positioning reference signal; the processor is configured to cause the device to send a reference signal transmission deactivation command; the device includes a location server, and wherein the processor is configured to cause the device to send a reference signal transmission activation command to one or more other devices configured to send a reference signal.

[0380] In further examples, according to examples disclosed herein, the processor 1904 and / or the transceiver 1908 may support wireless communications at the device 1902. For example, the processor 1904 and / or the transceiver 1908 may be configured to or otherwise support components for: sending one or more machine-learned positioning report requests, the machine-learned positioning report requests including a machine-learned reporting configuration and one or more common reporting criteria; receiving one or more machine-learned positioning reports; processing the one or more machine-learned positioning reports via a machine-learned model; and generating an estimated position of a user equipment (UE) based at least in part on an output from the machine-learned model.

[0381] In addition, in some implementations, the machine learning report configuration includes one or more of the following items: direct machine learning report configuration or assisted machine learning report configuration; the device includes a configuration entity, and wherein the configuration entity includes at least one of the following items: a location server, a next generation radio access network (NG-RAN), a positioning reference unit (PRU), a UE, an anchor UE, or a target UE; the processor is configured to cause the device to send one or more machine learning positioning report requests to one or more second devices, and receive one or more machine learning positioning reports from one or more second devices, and wherein the one or more second devices include at least one of the following items: a location server, a next generation radio access network (NG-RAN), a positioning reference unit (PRU), a UE, an anchor UE, or a target UE.

[0382] In addition, in some implementations, one or more public reporting standards include one or more of the following items: fingerprint type, ground truth reference position type, time domain reporting type, measurement preprocessing, measurement validity, UE type, mobility indication, orientation indication, fingerprint recognition environment indication, fingerprint quality indication or tag quality indication; for at least one machine learning positioning report request, one or more public reporting standards in the public reporting standards are configured to be added, removed, updated, activated or deactivated; the processor is configured to cause the device to broadcast the one or more public reporting standards via positioning system information broadcast signaling; the machine learning report configuration includes an indication of reporting machine learning positioning measurement correlation between different measurement sets.

[0383] In addition, in some implementations, the machine learning positioning measurement correlation includes one or more of the following: spatial correlation or time domain correlation; one or more machine learning positioning reports include one or more of the following: machine learning positioning measurements or machine learning positioning location information; the machine learning report configuration includes an indication to report path loss at different locations, the different locations including a ground truth reference location; the machine learning report configuration includes an indication to average the machine learning positioning measurements over a configured number of measurements and report the average as part of one or more machine learning positioning reports; the machine learning report configuration includes an indication to determine a similarity score at configured locations to determine an optimal mapping between the fingerprint measurement and the estimated location of the target UE.

[0384] In further examples, according to examples disclosed herein, the processor 1904 and / or the transceiver 1908 can support wireless communications at the device 1902. For example, the processor 1904 and / or the transceiver 1908 can be configured to or otherwise support components for: receiving one or more machine learning position report requests, the machine learning position report requests including a machine learning report configuration and one or more public reporting standards; generating one or more machine learning position reports based at least in part on the machine learning report configuration and the one or more public reporting standards; and sending the one or more machine learning position reports.

[0385] In addition, in some implementations, the apparatus includes one or more of the following: a user equipment (UE), an anchor UE, or a target UE; the apparatus includes a configuration entity, and wherein the configuration entity includes at least one of the following: a location server, a next generation radio access network (NG-RAN), or a positioning reference unit (PRU).

[0386] In further examples, according to examples disclosed herein, the processor 1904 and / or the transceiver 1908 can support wireless communications at the device 1902. For example, the processor 1904 and / or the transceiver 1908 can be configured to or otherwise support components for: sending one or more machine learning location report requests, the machine learning location report requests including a machine learning report configuration and one or more common reporting criteria; receiving one or more machine learning location reports; generating a machine learning location training dataset based at least in part on the one or more machine learning location reports; and training a location machine learning model using the machine learning location training dataset to generate a trained location machine learning model.

[0387] In addition, in some implementations, the processor and transceiver are configured to cause the device to: receive one or more additional machine learning positioning reports; input at least a portion of the one or more additional machine learning positioning reports into a trained positioning machine learning model; and estimate the location of a target user equipment (UE) based at least in part on the output from the trained positioning machine learning model; the processor is configured to cause the device to: receive a request for machine learning positioning training data for positioning; and based at least in part on the request, send a machine learning positioning training data set; the processor is configured to cause the device to: receive a request for a trained positioning machine learning model; and based at least in part on the request, send at least a portion of the trained positioning machine learning model.

[0388] In another example, according to examples disclosed herein, the processor 1904 and / or the transceiver 1908 can support wireless communications at the device 1902. For example, the processor 1904 and / or the transceiver 1908 can be configured to or otherwise support components for: receiving a machine learning positioning configuration request; and sending a machine learning positioning configuration response based at least in part on the machine learning positioning configuration request, the machine learning positioning configuration response including a positioning reference signal configuration to be measured at a corresponding reference location and utilizing one or more associated validity criteria.

[0389] In addition, in some implementations, for example, the processor 1904 and / or the transceiver 1908 may be configured to or otherwise support components for: receiving a machine learning positioning configuration request from one or more of a network node or a user equipment (UE) and sending a machine learning positioning configuration response to one or more of the network node or the UE; the reference location includes one or more ground truth reference locations; the machine learning positioning configuration response includes an artificial intelligence configuration for a positioning reference signal configuration; one or more machine learning positioning configuration responses are sent independently of the machine learning positioning configuration request; the machine learning positioning configuration response includes one or more of the following: a direct machine learning configuration or an assisted machine learning configuration; the direct machine learning configuration includes a configuration for performing radio frequency fingerprinting.

[0390] In addition, in some implementations, the machine learning positioning configuration request includes a request for one or more of the following: a training type, a ground truth position request, a machine learning method, a machine learning on-demand positioning reference signal request, or pre-configured machine learning assistance data; the positioning reference signal configuration includes one or more of the following: positioning reference signal resources, resource sets, transmission reception points, or positioning frequency layers to be measured at each reference location; one or more associated validity criteria include one or more of the following: a time standard or a spatial standard; the time standard includes a measurement time defined by one or more time bases; the one or more time bases include at least one of the following: a system frame number, coordinated universal time (UTC), or global navigation satellite system (GNSS) time; one or more associated validity criteria include a spatial standard relating to an indication of a geographic area, and the spatial standard includes one or more of the following: an area identifier, a cell identifier, or a segment identifier.

[0391] In addition, in some implementations, the positioning reference signal configuration includes one or more indications of one or more of the following items: uplink radio access technology-related measurements, downlink radio access technology-related measurements, or sidelink radio access technology-related measurements to be performed; the positioning reference signal configuration includes one or more indications of one or more of the following items: downlink radio access technology-independent measurements or sidelink radio access technology-independent measurements to be performed; the machine learning positioning configuration response also includes one or more configurations related to at least one of the following items: k-nearest neighbor (k-NN) algorithm, support vector machine, decision tree, random forest, Gaussian mixture model, Bayesian network, artificial neural network, k-means, reinforcement learning, deep learning, or transfer learning; the method is performed by an apparatus comprising at least one of the following items: a location server, a next generation radio access network (NG-RAN), a positioning reference unit (PRU), a user equipment (UE), an anchor UE, or a target UE.

[0392] In addition, in some implementations, the positioning reference signal configuration includes a received signal strength indicator measurement for one or more of the downlink, uplink, or sidelink; and the received signal strength indicator measurement for one or more of the downlink, uplink, or sidelink is defined as: comprising a linear average of the total received power observed in the resource elements of the time slot carrying the positioning reference signal configured for measurement; the positioning reference signal configuration includes a positioning reference signal arrival time measurement for one or more of the downlink, uplink, or sidelink; and the positioning reference signal arrival time measurement for one or more of the downlink, uplink, or sidelink is defined as comprising the reception time of the positioning reference signal at the receiver reference point; the positioning reference signal configuration includes an indication that the location server is permitted to send an error cause related to an incorrect configuration of the positioning reference signal configuration; the positioning reference signal configuration includes an indication that the target user equipment (UE) is permitted to send an error cause related to one or more of the following items: an error in receiving a machine learning positioning configuration response or a measurement error related to the machine learning positioning measurement.

[0393] In further examples, according to examples disclosed herein, the processor 1904 and / or the transceiver 1908 can support wireless communications at the device 1902. For example, the processor 1904 and / or the transceiver 1908 can be configured to or otherwise support components for: sending a machine learning position configuration request; receiving a machine learning position configuration response including a positioning reference signal configuration to be measured at one or more respective reference locations and utilizing one or more associated validity criteria; and performing one or more machine learning position measurements based at least in part on the positioning reference signal configuration.

[0394] In addition, in some implementations, for example, the processor 1904 and / or the transceiver 1908 may be configured to or otherwise support components for one or more of: receiving a machine learning positioning configuration response in response to a machine learning positioning configuration request; or receiving a machine learning positioning configuration response independent of the machine learning positioning configuration request; inputting one or more machine learning position measurements into a machine learning model and receiving output from the machine learning model; generating an estimated position of the device based at least in part on the output from the machine learning model; the method being performed by a device comprising one or more of: a target user equipment (UE), a positioning reference unit (PRU), or an anchor UE.

[0395] In another example, according to examples disclosed herein, the processor 1904 and / or the transceiver 1908 can support wireless communications at the device 1902. For example, the processor 1904 and / or the transceiver 1908 can be configured to or otherwise support components for: sending a configuration request to configure a reference signal for machine learning positioning measurements; receiving a configuration response including a reference signal configuration; and sending a reference signal transmission activation command based at least in part on the reference signal configuration.

[0396] In addition, in some implementations, the reference signal includes one or more of the following items: a sounding reference signal or a positioning reference signal; sending a reference signal transmission deactivation command; the method is performed by an apparatus, the apparatus including a location server, and wherein the method also includes sending a reference signal transmission activation command to one or more other apparatuses configured to send a reference signal.

[0397] In further examples, according to examples disclosed herein, the processor 1904 and / or the transceiver 1908 may support wireless communications at the device 1902. For example, the processor 1904 and / or the transceiver 1908 may be configured to or otherwise support components for: sending one or more machine-learned positioning report requests, the machine-learned positioning report requests including a machine-learned reporting configuration and one or more common reporting criteria; receiving one or more machine-learned positioning reports; processing the one or more machine-learned positioning reports via a machine-learned model; and generating an estimated position of a user equipment (UE) based at least in part on an output from the machine-learned model.

[0398] In addition, in some implementations, the machine learning report configuration includes one or more of the following items: direct machine learning report configuration or assisted machine learning report configuration; wherein the method is performed by an apparatus, the apparatus including a configuration entity, and wherein the configuration entity includes at least one of the following items: a location server, a next generation radio access network (NG-RAN), a positioning reference unit (PRU), a UE, an anchor UE, or a target UE; the method also includes: sending one or more machine learning positioning report requests from the first apparatus to one or more second apparatuses, and receiving one or more machine learning positioning reports from the one or more second apparatuses, and wherein the one or more second apparatuses include at least one of the following items: a location server, a next generation radio access network (NG-RAN), a positioning reference unit (PRU), a UE, an anchor UE, or a target UE.

[0399] In addition, in some implementations, one or more public reporting standards include one or more of the following items: fingerprint type, ground truth reference position type, time domain reporting type, measurement preprocessing, measurement validity, UE type, mobility indication, orientation indication, fingerprint recognition environment indication, fingerprint quality indication or tag quality indication; wherein for at least one machine learning positioning report request, one or more public reporting standards in the public reporting standards are configured to be added, removed, updated, activated or deactivated; the method also includes broadcasting one or more public reporting standards via positioning system information broadcast signaling; wherein the machine learning report configuration includes reporting an indication of the correlation of machine learning positioning measurements between different measurement sets.

[0400] In addition, in some implementations, the machine learning positioning measurement correlation includes one or more of the following: spatial correlation or time domain correlation; wherein the one or more machine learning positioning reports include one or more of the following: machine learning positioning measurements or machine learning positioning location information; wherein the machine learning report configuration includes an indication of reporting path loss at different locations, the different locations including a ground truth reference location; wherein the machine learning report configuration includes an indication of averaging the machine learning positioning measurements over a configured number of measurements and reporting the average as part of the one or more machine learning positioning reports; wherein the machine learning report configuration includes an indication of determining a similarity score at configured locations to determine an optimal mapping between the fingerprint measurement and the estimated location of the target UE.

[0401] In further examples, according to examples disclosed herein, the processor 1904 and / or the transceiver 1908 can support wireless communications at the device 1902. For example, the processor 1904 and / or the transceiver 1908 can be configured to or otherwise support components for: receiving one or more machine learning position report requests, the machine learning position report requests including a machine learning report configuration and one or more public reporting standards; generating one or more machine learning position reports based at least in part on the machine learning report configuration and the one or more public reporting standards; and sending the one or more machine learning position reports.

[0402] In addition, in some implementations, the method is performed by an apparatus comprising at least one of: a user equipment (UE), an anchor UE, or a target UE; wherein the method is performed by an apparatus comprising a configuration entity, and wherein the configuration entity comprises at least one of: a location server, a next generation radio access network (NG-RAN), or a positioning reference unit (PRU).

[0403] In further examples, according to examples disclosed herein, the processor 1904 and / or the transceiver 1908 can support wireless communications at the device 1902. For example, the processor 1904 and / or the transceiver 1908 can be configured as or otherwise used as a component for: sending one or more machine learning positioning report requests, the machine learning positioning report requests including a machine learning report configuration and one or more common reporting criteria; receiving one or more machine learning positioning reports; generating a machine learning positioning training dataset based at least in part on the one or more machine learning positioning reports; and training a positioning machine learning model using the machine learning positioning training dataset to generate a trained positioning machine learning model.

[0404] In addition, in some implementations, for example, the processor 1904 and / or the transceiver 1908 can be configured to or otherwise support components for: receiving one or more additional machine learning positioning reports; inputting at least a portion of the one or more additional machine learning positioning reports into a trained positioning machine learning model; and estimating the location of a target user equipment (UE) based at least in part on an output from the trained positioning machine learning model; the method also includes: receiving a request for machine learning positioning training data for positioning; and sending a machine learning positioning training data set based at least in part on the request; the method also includes: receiving a request for a trained positioning machine learning model; and sending at least a portion of the trained positioning machine learning model based at least in part on the request.

[0405] The processor 1904 may include an intelligent hardware device (e.g., a general-purpose processor, a DSP, a CPU, a microcontroller, an ASIC, an FPGA, a programmable logic device, discrete gate or transistor logic components, discrete hardware components, or any combination thereof). In some implementations, the processor 1904 may be configured to operate a memory array using a memory controller. In some other implementations, the memory controller may be integrated into the processor 1904. The processor 1904 may be configured to execute computer-readable instructions stored in a memory (e.g., memory 1906) to cause the device 1902 to perform various functions of the present disclosure.

[0406] The memory 1906 may include random access memory (RAM) and read-only memory (ROM). The memory 1906 may store computer-readable computer executable code, which includes instructions that, when executed by the processor 1904, cause the device 1902 to perform the various functions described herein. The code may be stored in a non-transitory computer-readable medium such as system memory or another type of memory. In some implementations, the code may not be directly executed by the processor 1904, but may cause a computer (e.g., when compiled and executed) to perform the functions described herein. In some implementations, the memory 1906 may include a basic I / O system (BIOS), etc., which may control basic hardware or software operations, such as interaction with peripheral components or devices.

[0407] I / O controller 1910 can manage input and output signals for device 1902. I / O controller 1910 can also manage peripheral devices that are not integrated into device 2402. In some implementations, I / O controller 1910 can represent a physical connection or port to an external peripheral device. In some implementations, I / O controller 1910 can utilize an operating system, such as MS or other known operating systems. In some implementations, I / O controller 1910 can be implemented as part of a processor such as processor M06. In some implementations, a user can interact with device 1902 via I / O controller 1910 or via hardware components controlled by I / O controller 1910.

[0408] In some implementations, the device 1902 may include a single antenna 1912. However, in some other implementations, the device 1902 may have more than one antenna 1912 (e.g., multiple antennas), including multiple antenna panels or antenna arrays, which are capable of concurrently sending or receiving multiple wireless transmissions. The transceiver 1908 can communicate bidirectionally via one or more antennas 1912, wired or wireless links, as described herein. For example, the transceiver 1908 can represent a wireless transceiver and can communicate bidirectionally with another wireless transceiver. The transceiver 1908 can also include a modem for modulating packets to provide the modulated packets to one or more antennas 1912 for transmission, and demodulating packets received from the one or more antennas 1912.

[0409] Figure 20 A flow chart of a method 2000 for supporting machine learning for positioning according to aspects of the present disclosure is illustrated. The operations of the method 2000 may be implemented by the devices described herein or components thereof. For example, the operations of the method 2000 may be implemented by reference to Figures 1 to 19The network entity 102 and / or UE 104 performs the described functions. In some implementations, the device may execute a set of instructions to control the functional elements of the device to perform the described functions. Additionally or alternatively, the device may use dedicated hardware to perform aspects of the described functions.

[0410] At 2002, the method may include receiving a machine learning positioning configuration request. The operations of 2002 may be performed according to the examples described herein. In some implementations, aspects of the operations of 2002 may be described by reference to Figure 1 The device is used to perform the above.

[0411] At 2004, the method may include: sending a machine learning positioning configuration response based at least in part on the machine learning positioning configuration request, the machine learning positioning configuration response including a positioning reference signal configuration to be measured at a corresponding reference position and utilizing one or more associated validity criteria. The operations of 2004 may be performed according to the examples described herein. In some implementations, aspects of the operations of 2004 may be performed by reference to Figure 1 The device is used to perform the above.

[0412] Figure 21 A flow chart of a method 2100 for supporting machine learning for positioning according to aspects of the present disclosure is illustrated. The operations of the method 2100 may be implemented by the devices described herein or components thereof. For example, the operations of the method 2100 may be implemented by reference to Figures 1 to 19 The network entity 102 and / or UE 104 performs the described functions. In some implementations, the device may execute a set of instructions to control the functional elements of the device to perform the described functions. Additionally or alternatively, the device may use dedicated hardware to perform aspects of the described functions.

[0413] At 2102, the method may include sending a machine learning positioning configuration request. The operations of 2102 may be performed according to the examples described herein. In some implementations, aspects of the operations of 2102 may be described by reference to Figure 1 The device is used to perform the above.

[0414] At 2104, the method may include receiving a machine learning positioning configuration response including a positioning reference signal configuration to be measured at one or more corresponding reference locations and using one or more associated validity criteria. The operations of 2104 may be performed according to examples described herein. In some implementations, aspects of the operations of 2104 may be performed by reference to Figure 1 The device is used to perform the above.

[0415] At 2106, the method may include performing one or more machine learning position measurements based at least in part on the positioning reference signal configuration. The operations of 2106 may be performed according to the examples described herein. In some implementations, aspects of the operations of 2106 may be performed by reference to Figure 1 The device is used to perform the above.

[0416] Figure 22 A flow chart of a method 2200 for supporting machine learning for positioning according to aspects of the present disclosure is illustrated. The operations of the method 2200 may be implemented by the devices described herein or components thereof. For example, the operations of the method 2200 may be implemented by reference to Figures 1 to 19 The network entity 102 and / or UE 104 performs the described functions. In some implementations, the device may execute a set of instructions to control the functional elements of the device to perform the described functions. Additionally or alternatively, the device may use dedicated hardware to perform aspects of the described functions.

[0417] At 2202, the method may include sending a configuration request to configure a reference signal for machine learning positioning measurements. The operations of 2202 may be performed according to the examples described herein. In some implementations, aspects of the operations of 2202 may be performed by reference to Figure 1 The device is used to perform the above.

[0418] At 2204, the method may include receiving a configuration response including a reference signal configuration. The operations of 2204 may be performed according to examples described herein. In some implementations, aspects of the operations of 2204 may be performed by reference signal configuration. Figure 1 The device is used to perform the above.

[0419] At 2206, the method may include: sending a reference signal transmission activation command based at least in part on the reference signal configuration. The operations of 2206 may be performed according to the examples described herein. In some implementations, aspects of the operations of 2206 may be performed by reference signal configuration. Figure 1 The device is used to perform the above.

[0420] Figure 23 A flow chart of a method 2300 for supporting machine learning for positioning according to aspects of the present disclosure is illustrated. The operations of the method 2300 may be implemented by the devices described herein or components thereof. For example, the operations of the method 2300 may be implemented by reference to Figures 1 to 19 The network entity 102 and / or UE 104 performs the described functions. In some implementations, the device may execute a set of instructions to control the functional elements of the device to perform the described functions. Additionally or alternatively, the device may use dedicated hardware to perform aspects of the described functions.

[0421] At 2302, the method may include sending one or more machine learning positioning report requests, the machine learning positioning report requests including a machine learning reporting configuration and one or more public reporting standards. The operations of 2302 may be performed according to the examples described herein. In some implementations, aspects of the operations of 2302 may be described with reference to Figure 1 The device is used to perform the above.

[0422] At 2304, the method may include receiving one or more machine learning positioning reports. The operations of 2304 may be performed according to the examples described herein. In some implementations, aspects of the operations of 2304 may be described with reference to Figure 1 The device is used to perform the above.

[0423] At 2306, the method may include processing one or more machine learning positioning reports via a machine learning model. The operations of 2306 may be performed according to the examples described herein. In some implementations, aspects of the operations of 2306 may be described with reference to Figure 1 The device is used to perform the above.

[0424] At 2308, the method may include generating an estimated location of a user equipment (UE) based at least in part on the output from the machine learning model. The operations of 2308 may be performed according to the examples described herein. In some implementations, aspects of the operations of 2308 may be described with reference to Figure 1 The device is used to perform the above.

[0425] Figure 24 A flow chart of a method 2400 for supporting machine learning for positioning according to aspects of the present disclosure is illustrated. The operations of the method 2400 may be implemented by the devices described herein or components thereof. For example, the operations of the method 2400 may be implemented by reference to Figures 1 to 19 The network entity 102 and / or UE 104 performs the described functions. In some implementations, the device may execute a set of instructions to control the functional elements of the device to perform the described functions. Additionally or alternatively, the device may use dedicated hardware to perform aspects of the described functions.

[0426] At 2402, the method may include receiving one or more machine learning positioning report requests, the machine learning positioning report requests including a machine learning reporting configuration and one or more public reporting standards. The operations of 2402 may be performed according to the examples described herein. In some implementations, aspects of the operations of 2402 may be performed by reference to Figure 1 The device is used to perform the above.

[0427] At 2404, the method may include generating one or more machine learning positioning reports based at least in part on a machine learning reporting configuration and one or more public reporting standards. The operations of 2404 may be performed according to examples described herein. In some implementations, aspects of the operations of 2404 may be described with reference to Figure 1 The device is used to perform the above.

[0428] At 2406, the method may include sending one or more machine learning positioning reports. The operations of 2406 may be performed according to the examples described herein. In some implementations, aspects of the operations of 2406 may be described with reference to Figure 1 The device is used to perform the above.

[0429] Figure 25 A flow chart of a method 2500 for supporting machine learning for positioning according to aspects of the present disclosure is illustrated. The operations of the method 2500 may be implemented by the devices described herein or components thereof. For example, the operations of the method 2500 may be implemented by reference to Figures 1 to 19 The network entity 102 and / or UE 104 performs the described functions. In some implementations, the device may execute a set of instructions to control the functional elements of the device to perform the described functions. Additionally or alternatively, the device may use dedicated hardware to perform aspects of the described functions.

[0430] At 2502, the method may include sending one or more machine learning positioning report requests, the machine learning positioning report requests including a machine learning reporting configuration and one or more public reporting standards. The operations of 2502 may be performed according to the examples described herein. In some implementations, aspects of the operations of 2502 may be described with reference to Figure 1 The device is used to perform the above.

[0431] At 2504, the method may include receiving one or more machine learning positioning reports. The operations of 2504 may be performed according to the examples described herein. In some implementations, aspects of the operations of 2504 may be described with reference to Figure 1 The device is used to perform the above.

[0432] At 2506, the method may include generating a machine learning positioning training data set based at least in part on the one or more machine learning positioning reports. The operations of 2506 may be performed according to the examples described herein. In some implementations, aspects of the operations of 2506 may be described with reference to Figure 1 The device is used to perform the above.

[0433] At 2508, the method may include: training the positioning machine learning model using the machine learning positioning training dataset to generate a trained positioning machine learning model. The operation of 2508 may be performed according to the examples described herein. In some implementations, aspects of the operation of 2508 may be described by reference to Figure 1 The device is used to perform the above.

[0434] It should be noted that the methods described herein describe possible implementations, and that the operations and steps may be rearranged or otherwise modified, and other implementations are possible. Furthermore, aspects from two or more methods may be combined.

[0435] The various illustrative blocks and components described in conjunction with the disclosure herein may be implemented or executed using a general purpose processor, a DSP, an ASIC, a CPU, an FPGA or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A general purpose processor may be a microprocessor, but in the alternative, the processor may be any processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of computing devices (e.g., a DSP and a microprocessor, a plurality of microprocessors, a combination of one or more microprocessors in conjunction with a DSP core, or any other such configuration).

[0436] The functions described herein may be implemented in hardware, software executed by a processor, firmware, or any combination thereof. If implemented in software executed by a processor, the functions may be stored on or transmitted via a computer-readable medium as one or more instructions or code. Other examples and implementations are within the scope of this disclosure and the appended claims. For example, due to the nature of software, the functions described herein may be implemented using software executed by a processor, hardware, firmware, hardwiring, or any combination thereof. Features that implement the functions may also be physically located in various locations, including being distributed so that portions of the functions are implemented at different physical locations.

[0437] Computer readable medium includes non-transient computer storage medium and communication medium, and communication medium includes any medium that promotes that computer program is transferred from one place to another place.Non-transient storage medium can be any available medium that can be accessed by general-purpose or special-purpose computer.As an example and not limitation, non-transient computer readable medium can include RAM, ROM, electrically erasable programmable ROM (EEPROM), flash memory, compact disc (CD) ROM or other optical disc storage device, magnetic disk storage device or other magnetic storage device, or can be used for carrying or storing desired program code components in the form of instruction or data structure and can be by general-purpose or special-purpose computer, or any other non-transient medium that general-purpose or special-purpose processor accesses.

[0438] Any connection may be appropriately referred to as a computer-readable medium. For example, if the software is sent from a website, server, or other remote source using a coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technology (such as infrared, radio, and microwave), the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technology (such as infrared, radio, and microwave) is included in the definition of computer-readable medium. Disks and optical discs as used herein include CDs, laser discs, optical discs, digital versatile discs (DVDs), floppy disks, and Blu-ray discs, where disks typically reproduce data magnetically, while optical discs reproduce data optically using lasers. Combinations of the above are also included within the scope of computer-readable media.

[0439] As used herein (including in the claims), "or" used in a list of items (e.g., a list of items beginning with a phrase such as "at least one of" or "one or more of" or "one or both of") represents an inclusive list, so that, for example, a list of at least one of A, B, or C represents A or B or C or AB or AC or BC or ABC (e.g., A and B and C). Furthermore, as used herein, the phrase "based on" should not be interpreted as a reference to a set of closed conditions. For example, an example step described as "based on condition A" can be based on both condition A and condition B without departing from the scope of this disclosure. In other words, as used herein, the phrase "based on" should be interpreted in the same manner as the phrase "based at least in part on." Furthermore, as used herein (including in the claims), a "set" may include one or more elements.

[0440] When referring to a network entity, the terms "send," "receive," or "transmit" may refer to any part of a network entity of the RAN (e.g., base station, CU, DU, RU) that communicates with another device (e.g., directly or via one or more other network entities).

[0441] The description herein in conjunction with the accompanying drawings describes example configurations and does not represent all examples that can be implemented or within the scope of the claims. The term "example" as used herein means "used as an example, instance, or illustration," rather than "preferred" or "superior to other examples." The detailed description includes specific details to facilitate understanding of the described techniques. However, these techniques can be practiced without these specific details. In some cases, known structures and devices are shown in block diagram form to avoid obscuring the concepts of the described examples.

[0442] The description herein is provided to enable one of ordinary skill in the art to make or use the present disclosure. Various modifications to the present disclosure will be readily apparent to one of ordinary skill in the art, and the general principles defined herein may be applied to other variations without departing from the scope of the present disclosure. Therefore, the present disclosure is not limited to the examples and designs described herein, but should be construed in the widest sense consistent with the principles and novel features disclosed herein.

Claims

1. A device comprising: processor; as well as a memory coupled to the processor, the processor being configured to cause the apparatus to: sending one or more machine learning positioning report requests, the machine learning positioning report requests including a machine learning reporting configuration and one or more common reporting criteria; receiving one or more machine learning positioning reports; processing the one or more machine learning location reports via a machine learning model; as well as An estimated location of a user equipment (UE) is generated based at least in part on the output from the machine learning model.

2. The apparatus of claim 1 , wherein the machine learning reporting configuration comprises one or more of: a direct machine learning reporting configuration, or an assisted machine learning reporting configuration.

3. The apparatus of claim 1 , wherein the apparatus comprises a configuration entity, and wherein the configuration entity comprises at least one of the following: a location server, a next generation radio access network (NG-RAN), a positioning reference unit (PRU), a UE, an anchor UE, or a target UE.

4. The apparatus of claim 1 , wherein the processor is configured to cause the apparatus to send the one or more machine learning positioning report requests to one or more second apparatuses and receive the one or more machine learning positioning reports from the one or more second apparatuses, and wherein the one or more second apparatuses include at least one of the following: a location server, a next generation radio access network (NG-RAN), a positioning reference unit (PRU), a UE, an anchor UE, or a target UE.

5. The apparatus of claim 1 , wherein the one or more common reporting criteria comprise one or more of: fingerprint type, ground truth reference location type, time domain reporting type, pre-processing of measurements, measurement validity, UE type, mobility indication, orientation indication, fingerprinting environment indication, fingerprint quality indication, or tag quality indication.

6. The apparatus of claim 1 , wherein for at least one machine learning location reporting request, the one or more of the public reporting standards are configured to be one or more of enhanced, removed, updated, activated, or deactivated. 7 . The apparatus of claim 1 , wherein the processor is configured to cause the apparatus to broadcast the one or more common reporting standards via positioning system information broadcast signaling.

8. The apparatus of claim 1 , wherein the machine learning report configuration comprises: An indication to report the correlation of machine learning positioning measurements between different measurement sets.

9. The apparatus of claim 8, wherein the machine learning positioning measurement correlation comprises one or more of: spatial domain correlation, or temporal domain correlation.

10. The apparatus of claim 1, wherein the one or more machine learning positioning reports comprise one or more of: machine learning positioning measurements, or machine learning positioning position information.

11. The apparatus of claim 1 , wherein the machine learning reporting configuration includes an indication to report path loss at different locations, the different locations comprising a ground truth reference location.

12. The apparatus of claim 1 , wherein the machine learning report configuration comprises: Instructions to average machine-learned positioning measurements over a configured number of measurements and report the average as part of the one or more machine-learned positioning reports.

13. The apparatus of claim 1 , wherein the machine learning report configuration comprises: An indication of how to determine a similarity score at the configured positions to determine an optimal mapping between the fingerprint measurements and the estimated position of the target UE.

14. An apparatus comprising: processor; as well as a memory coupled to the processor, the processor being configured to cause the apparatus to: receiving one or more machine learning positioning report requests, the machine learning positioning report requests comprising a machine learning reporting configuration and one or more common reporting criteria; generating one or more machine learning positioning reports based at least in part on the machine learning reporting configuration and the one or more common reporting standards; as well as Send the one or more machine learning positioning reports.

15. The apparatus of claim 14, wherein the apparatus comprises one or more of: a user equipment (UE), an anchor UE, or a target UE.

16. The apparatus of claim 14, wherein the apparatus comprises a configuration entity, and wherein the configuration entity comprises at least one of: a location server, a next generation radio access network (NG-RAN), or a positioning reference unit (PRU).

17. An apparatus comprising: processor; as well as a memory coupled to the processor, the processor being configured to cause the apparatus to: sending one or more machine learning positioning report requests, the machine learning positioning report requests including a machine learning reporting configuration and one or more common reporting criteria; receiving one or more machine learning positioning reports; generating a machine learning positioning training dataset based at least in part on the one or more machine learning positioning reports; as well as The positioning machine learning model is trained using the machine learning positioning training dataset to generate a trained positioning machine learning model.

18. The apparatus of claim 17, wherein the processor is configured to cause the apparatus to: receiving one or more additional machine learning positioning reports; inputting at least a portion of the one or more additional machine-learned positioning reports into the trained positioning machine-learning model; and A location of a target user equipment (UE) is estimated based at least in part on output from the trained positioning machine learning model.

19. The apparatus of claim 17, wherein the processor is configured to cause the apparatus to: receiving a request for machine learning positioning training data for positioning; and Based at least in part on the request, the machine learning positioning training dataset is sent.

20. The apparatus of claim 17, wherein the processor is configured to cause the apparatus to: receiving a request for a trained positioning machine learning model; and Based at least in part on the request, at least a portion of the trained positioning machine learning model is sent.

Images