Updating machine learning models for positioning

Abstract

A method is disclosed, the method including: obtaining a machine learning model for positioning, the machine learning model being trained based on first training data associated with one or more first network nodes; transmitting information including at least one of the machine learning model, a request to update the machine learning model based on second training data, or a request to provide second training data for updating the machine learning model at the device; receiving at least one of a message indicating the updated machine learning model or the second training data; and transmitting the updated machine learning model to at least one or more second network nodes.

Description

Technical Field

[0001] The following exemplary embodiments relate to updating a machine learning model for wireless communication and positioning.

Background Art

[0002] Positioning technology may be used to estimate the physical location of a device. It is desirable to improve positioning accuracy to more accurately estimate the location of the device.

Summary of the Invention

[0003] The scope of protection sought for various exemplary embodiments is described by the independent claims. Exemplary embodiments and features described herein that do not fall within the scope of the independent claims, if any, are to be construed as useful examples for understanding the various embodiments.

[0004] According to one aspect, there is provided an apparatus comprising at least one processor and at least one memory storing instructions which, when executed by the at least one processor, cause the apparatus to, at least, obtain a machine learning model for positioning, wherein the machine learning model is trained based on first training data associated with one or more first network nodes, obtain; transmit information including at least one of a request to update the machine learning model based on the machine learning model and second training data, or a request to provide second training data for updating the machine learning model in the apparatus; receive at least one of a message indicating the updated machine learning model or the second training data; and transmit the updated machine learning model to one or more second network nodes.

[0005] In another embodiment, an apparatus is provided, the apparatus comprising means for acquiring a machine learning model for positioning, the machine learning model being trained on first training data associated with one or more first network nodes; means for transmitting information including at least one of a machine learning model, a request to update the machine learning model based on second training data, or a request for the apparatus to provide second training data for updating the machine learning model; means for receiving at least one of a message indicating an updated machine learning model or second training data; and means for transmitting the updated machine learning model to at least one or more second network nodes.

[0006] In another embodiment, a method is provided, the method comprising acquiring a machine learning model for positioning by a device, the machine learning model being trained on first training data associated with one or more first network nodes; transmitting information by the device including at least one of a machine learning model, a request to update the machine learning model based on second training data, or a request for the device to provide second training data for updating the machine learning model; receiving at least one of a message indicating an updated machine learning model or second training data by the device; and transmitting the updated machine learning model by the device to at least one or more second network nodes.

[0007] In another embodiment, a computer program including instructions is provided, and when executed by a device, the instructions cause the device to perform at least: acquire a machine learning model for positioning, the machine learning model being trained on first training data associated with one or more first network nodes; transmit information including at least one of a machine learning model, a request to update the machine learning model based on second training data, or a request to provide the device with second training data for updating the machine learning model; receive at least one of a message indicating an updated machine learning model or second training data; and transmit the updated machine learning model to at least one or more second network nodes.

[0008] In another embodiment, a computer-readable medium containing program instructions is provided, and when executed by a device, the program instructions cause the device to perform at least the following: acquire a machine learning model for positioning, the machine learning model being trained on first training data associated with one or more first network nodes; transmit information including at least one of a machine learning model, a request to update the machine learning model based on second training data, or a request to provide second training data for updating the machine learning model in the device; receive at least one of a message indicating an updated machine learning model or second training data; and transmit the updated machine learning model to at least one or more second network nodes.

[0009] In another embodiment, a non-temporary computer-readable medium containing program instructions is provided, and when executed by the device, the program instructions cause the device to perform at least: acquire a machine learning model for positioning, the machine learning model being trained on first training data associated with one or more first network nodes; transmit information including at least one of a machine learning model, a request to update the machine learning model based on second training data, or a request to provide the device with second training data for updating the machine learning model; receive at least one of a message indicating the updated machine learning model or second training data; and transmit the updated machine learning model to at least one or more second network nodes.

[0010] In another embodiment, a device is provided comprising at least one processor and at least one memory for storing instructions, wherein when the instruction is executed by at least one processor, the device is caused to receive information comprising at least one of: a machine learning model for positioning; a request to update the machine learning model in the device based on second training data; or a request to provide second training data for updating the machine learning model, wherein the machine learning model is trained based on first training data associated with one or more first network nodes; to receive; to acquire second training data; and to transmit at least one of: a message indicating the updated machine learning model; or second training data.

[0011] In another embodiment, a device is provided, the device comprising means for receiving information including at least one of a machine learning model for positioning, a request to update the machine learning model in the device based on second training data, or a request to provide second training data for updating the machine learning model, wherein the machine learning model is trained based on first training data associated with one or more first network nodes, means for acquiring second training data, and means for transmitting at least one of a message indicating an updated machine learning model, or second training data.

[0012] In another embodiment, a method is provided, the method comprising: receiving information by the device including at least one of a machine learning model for positioning, a request for updating the machine learning model in the device based on second training data, or a request for providing second training data for updating the machine learning model, wherein the machine learning model is trained based on first training data associated with one or more first network nodes; obtaining second training data by the device; and transmitting at least one of a message indicating the updated machine learning model, or second training data by the device.

[0013] In another embodiment, a computer program is provided which, when executed by a device, causes the device to receive information including at least one of: a machine learning model for positioning; a request to update the machine learning model in the device based on second training data; or a request to provide second training data for updating the machine learning model, wherein the machine learning model is trained based on first training data associated with one or more first network nodes; to receive; to acquire second training data; and to transmit at least one of: a message indicating an updated machine learning model; or second training data.

[0014] In another embodiment, a computer-readable medium containing program instructions is provided, and when executed by a device, the program instructions cause the device to receive information including at least one of: a machine learning model for positioning; a request to update the machine learning model in the device based on second training data; or a request to provide second training data for updating the machine learning model, wherein the machine learning model is trained based on first training data associated with one or more first network nodes; receive; acquire second training data; and transmit at least one of: a message indicating an updated machine learning model; or second training data.

[0015] In another embodiment, a non-temporary computer-readable medium containing program instructions is provided, and when executed by a device, the program instructions cause the device to receive information including at least one of: a machine learning model for positioning; a request to update the machine learning model in the device based on second training data; or a request to provide second training data for updating the machine learning model, wherein the machine learning model is trained based on first training data associated with one or more first network nodes; receive; acquire second training data; and transmit at least one of: a message indicating an updated machine learning model; or second training data.

[0016] Various exemplary embodiments will be described in more detail below with reference to the accompanying drawings. [Brief explanation of the drawing]

[0017] [Figure 1] This is a diagram showing an example of a cellular communication network. [Figure 2] This figure shows an example of a positioning scenario. [Figure 3] This is a signal transmission diagram according to an exemplary embodiment. [Figure 4]This is a signal transmission diagram according to an exemplary embodiment. [Figure 5] This is a flowchart illustrating an exemplary embodiment. [Figure 6] This is a flowchart illustrating an exemplary embodiment. [Figure 7] This is a diagram showing an example of the device. [Figure 8] This is a diagram showing an example of the device. [Figure 9] This is a diagram showing an example of the device. [Figure 10] This figure shows an example of an artificial neural network. [Figure 11] This figure shows an example of a computing node. [Modes for carrying out the invention]

[0018] The following embodiments are illustrative. While this specification may refer to “an,” “one,” or “some” embodiments in several locations of the text, this does not necessarily mean that each reference is made to the same embodiment or that a particular feature applies only to a single embodiment. A single feature from different embodiments may be combined to provide other embodiments.

[0019] In the following, different exemplary embodiments are described using a radio access architecture based on Long Term Evolution Advanced (LTE-A), New Radio (NR, 5G), beyond 5G or 6th generation (6G) as examples of access architectures to which the exemplary embodiments may be applied. However, the exemplary embodiments are described without limiting them to such architectures. It will be apparent to those skilled in the art that the exemplary embodiments may be applied to other types of communication networks having suitable means by appropriately adjusting parameters and procedures. Some examples of other options for a suitable system are Universal Mobile Telecommunications System (UMTS) radio access network (UTRAN or E-UTRAN), Long Term Evolution (LTE, substantially the same as E-UTRAN), Wireless Local Area Network (WLAN or Wi-Fi), Worldwide Interoperability for Microwave Access (WiMAX), Bluetooth (registered trademark), Personal Communications Service (PCS), ZigBee (registered trademark), Wideband Code Division Multiple Access (WCDMA), systems using Ultra-Wideband (UWB) technology, sensor networks, Mobile Ad-Hoc Network (MANET), and Internet Protocol Multimedia Subsystem (IMS), or any combination thereof.

[0020] FIG. 1 shows an example of a simplified system architecture that includes several elements and functional entities, all of which are logical units, and may differ from that shown in its implementation. The connections shown in FIG. 1 are logical connections, and the actual physical connections may be different. It will be apparent to those skilled in the art that the system may have functions and structures other than those shown in FIG. 1.

[0021] Exemplary embodiments, however, are not limited to the systems given as examples, and those skilled in the art may apply solutions to other communication systems with the required characteristics.

[0022] The example of FIG. 1 shows a part of an exemplary radio access network.

[0023] FIG. 1 shows an access node (AN) 104, such as an evolved Node B (abbreviated as eNB or eNodeB) or a next generation Node B (abbreviated as gNB or gNodeB), which provides a radio cell, and user devices 100 and 102 configured to be in a wireless connection state with respect to one or more communication channels within the radio cell. The physical link from the user device to the access node may be called an uplink (UL) or reverse link, and the physical link from the access node to the user device may be called a downlink (DL) or forward link. The user devices may communicate directly with another user device by sidelink (SL) communication. It should be recognized that the access node or their functions may be performed by using an entity suitable for such use, such as any node, host, server, or access point.

[0024] The communication system may have two or more access nodes, in which case the access nodes may be configured to communicate with each other via links (wired or wireless) designed for that purpose. These links may be used for signal transmission and, similarly, for routing data from one access node to another. An access node may be a computing device configured to control the radio resources of the communication system to which the access nodes are coupled. An access node may also be called a base station, a base transceiver station (BTS), an access point, or any other type of interfacing device, including a relay station capable of operating in a wireless environment. An access node may include or be coupled to a transceiver. From the transceiver of the access node, a connection may be provided to an antenna unit that establishes a bidirectional radio link to a user device. The antenna unit may have multiple antennas or antenna elements. The access nodes may be further connected to a core network 110 (CN (core network) or next generation core NGC (NGC)). Depending on the technology deployed, one of the pair of access nodes that may be connected on the CN side may be a serving gateway (S-GW, which routes and forwards user data packets), a packet data network gateway (P-GW), etc., to provide connectivity for user devices to external packet data networks, user plane functions (UPF), mobility management entities (MME), or access and mobility management functions (AMF).

[0025] With regard to positioning, the service-based architecture (core network) may include an AMF 111 and a location management function (LMF) 112. The AMF may provide location information for call processing, policies, and billing to other network functions within the core network and other entities requesting positioning of terminal devices. The AMF may receive and manage location requests from several sources, namely mobile-originated location requests (MO-LR) from user devices and mobile-terminated location requests (MT-LR) from other functions or other network elements of the core network. The AMF may select an LMF for a given request and use its positioning service to trigger a positioning session. The LMF may then perform positioning upon receiving such a request from the AMF. The LMF may manage the resources and timing of positioning activities. The LMF may use the Namf_Communication service on the NL1 interface to request positioning of a user device from one or more access nodes, or the LMF may communicate with the user device through N1 for UE-based or UE-assisted positioning. Positioning may include location estimation, and the LMF may further estimate the movement or accuracy of location information if requested. In relation to connectivity, the AMF is located between the access node and the LMF and may therefore be closer to the access node than the LMF.

[0026] A user device represents one type of device to which resources on an air interface may be allocated and assigned; therefore, any feature described herein with respect to a user device may be performed using a corresponding device such as a relay node.

[0027] An example of such a relay node may be a Layer 3 relay to an access node (a self-backhauling relay). A self-backhauling relay node may also be called an integrated access and backhaul (IAB) node. An IAB node may comprise two ethical parts: a mobile termination (MT) part that undertakes the backhaul link (i.e., the link between the IAB node and a donor node also known as a parent node), and a distributed unit (DU) part that undertakes the access link (i.e., the child link between the IAB node and a user device and / or between the IAB node and other IAB nodes, a multi-hop scenario).

[0028] Another example of such a relay node may be a Layer 1 relay called a repeater. The repeater may amplify the signal received from the access node and forward the signal to the user device, and / or amplify the signal received from the user device and forward the signal to the access node.

[0029] User devices may be referred to by several names or names, including subscriber units, mobile stations, remote terminals, access terminals, user terminals, terminal devices, or user equipment (UE). User devices may also refer to portable computing devices, including wireless mobile communication devices that operate with or without a subscriber identification module (SIM), and portable computing devices include, but are not limited to, devices of the following types: mobile stations (mobile phones), smartphones, personal digital assistants (PDAs), handsets, devices using wireless modems (such as alarm or measuring devices), laptops, and / or touchscreen computers, tablets, game consoles, notebooks, multimedia devices, reduced capability (RedCap) devices, wireless sensor devices, or any devices integrated into vehicles.

[0030] It should be recognized that a user device may be a nearly exclusive uplink-only device, for example, a camera or video camera that loads images or video clips onto the network. A user device may be a device capable of operating in an Internet of Things (IoT) network, a scenario in which objects may have the ability to transfer data over the network without requiring human-to-human or human-to-computer interaction. A user device may utilize the cloud. In some applications, a user device may comprise a small, portable or wearable device with wireless components (such as a watch, earphones, or glasses), and computation may be performed in the cloud or on another user device. A user device (or, in some exemplary embodiments, a Layer 3 relay node) may be configured to perform one or more user device functions.

[0031] The various techniques described herein may be applied to cyber-physical systems (CPS) (systems that bring together computational elements to control physical entities). CPS may enable the execution and utilization of a vast number of interconnected ICT devices (sensors, actuators, processors, microcontrollers, etc.) embedded in physical objects at different locations. Mobile cyber-physical systems, where the physical system in question may have its own mobility, are a subcategory of cyber-physical systems. Examples of mobile-physical systems include mobile robotics and electronics transported by humans or animals.

[0032] Furthermore, although the device is shown as a single entity, different units, processors, and / or memory units (not all of which are shown in Figure 1) may be running.

[0033] 5G enables more base stations or nodes than LTE (the so-called small cell concept), including macrosites that work in cooperation with smaller stations using multiple input-multiple output (MIMO) antennas and employing various radio technologies depending on service needs, use cases, and / or available spectrum. 5G mobile communications may support a wide range of use cases and related applications, including video streaming, augmented reality, different methods of data sharing, and various forms of machine-type applications (such as massive machine-type communication (mMTC) including vehicle safety, different sensors, and real-time control). 5G may have multiple radio interfaces, i.e., sub-6GHz, cm wave, and mm wave, and also be integrable with existing legacy radio access technologies such as LTE. Integration with LTE may be implemented at least in the early phases as a system where macro coverage may be provided by LTE, and 5G radio interface access may originate from small cells by aggregation to LTE. In other words, 5G may support both RAT-to-RAT operability (such as LTE-5G) and RI-to-RI operability (operability between radio interfaces such as sub-6GHz-cm wave-mm wave). One concept that may be used in a 5G network is network slicing, in which multiple independent and dedicated virtual subnetworks (network instances) may be created within substantially the same infrastructure to run services with different requirements regarding latency, reliability, throughput, and mobility.

[0034] The current architecture in LTE networks may be fully distributed wirelessly and fully centralized in the core network. Low-latency applications and services in 5G may require bringing content closer to the wireless, which leads to local breakout and multi-access edge computing (MEC). 5G may enable analytics and knowledge generation to occur at the data source. This approach may require making effective use of resources that may not be continuously connected to the network, such as laptops, smartphones, tablets, and sensors. MEC may provide a distributed computing environment for application and service hosting. MEC may have the ability to store and process content in close proximity to cellular subscribers for faster response times. Edge computing may cover a wide range of technologies, including wireless sensor networks, mobile data acquisition, mobile signature analysis, collaborative distributed peer-to-peer ad-hoc networking, as well as cloud / fog computing and grid / mesh computing, due computing, mobile edge computing, cloudlets, distributed data storage and retrieval, autonomous self-healing networks, remote cloud services, augmented and virtual reality, data caching, the Internet of Things (massive connectivity and / or latency critical), and critical communications (autonomous vehicles, traffic safety, real-time analytics, time-critical control, healthcare applications).

[0035] The communication system may be able to communicate with or utilize services provided by one or more other networks 113, such as a public switched telephone network or the Internet. The communication network may also be able to support the use of cloud services, for example, at least part of the core network operation may be implemented as a cloud service (as shown in Figure 1 by “cloud” 114). The communication system may include a central control entity or similar which provides facilities for different operators’ networks to collaborate, for example, in spectrum sharing.

[0036] An access node may be divided into radio transceivers (TRX), i.e., radio units (RU) comprising a transmitter (Tx) and a receiver (Rx), one or more distributed units (DU) 105 which may be used for so-called Layer 1 (L1) processing and real-time Layer 2 (L2) processing, and a central unit (CU) 108 (also known as a centralized unit) which may be used for non-real-time L2 and Layer 3 (L3) processing. The CU 108 may be connected to one or more DU 105, for example, via an F1 interface. Such a division may allow for the centralization of CUs relative to cell sites and DUs, while DUs may be more distributed and remain further at cell sites. Both CUs and DUs may be called baseband or baseband units (BBUs). CUs and DUs may be contained within a radio access point (RAP).

[0037] CU108 may be defined as a logical node that hosts higher-layer protocols for the access node, such as radio resource control (RRC), service data adaptation protocol (SDAP), and / or packet data convergence protocol (PDCP). DU105 may be defined as a logical node that hosts radio link control (RLC), medium access control (MAC), and / or physical (PHY) layer for the access node. The operation of the DU may be controlled at least partially by the CU. The CU may comprise a control plane (CU-CP) which may be defined as a logical node hosting RRC, and the control plane portion of the CU's PDCP protocol for the access node. The CU may further comprise a user plane (CU-UP) which may be defined as a logical node hosting the user plane portions of the CU's PDCP and SDAP protocols for the access node.

[0038] A cloud computing platform may be used to run CU108 and / or DU105. The CU may run on the cloud computing platform, which may be called a virtualized CU (vCU). In addition to the vCU, there may be a virtualized DU (vDU) running on the cloud computing platform. Furthermore, there may be combinations in which the DU uses a so-called bare-metal solution, e.g., an application-specific integrated circuit (ASIC) or a customer-specific standard product (CSSP) system-on-a-chip (SoC) solution. It should also be understood that the distribution of functions between the access node units described above, or between different core network operations and access node operations, may differ.

[0039] An edge cloud may be made to participate in a radio access network (RAN) by utilizing network function virtualization (NFV) and software-defined networking (SDN). Using an edge cloud may mean that access node operations are performed at least partially on servers, hosts, or nodes that are operably coupled to access nodes having a remote radio head (RRH) or radio unit (RU), or radio portion. Node operations may also be distributed across multiple servers, nodes, or hosts. Applications of a cloud RAN architecture enable RAN real-time functions to be performed on the RAN side (e.g., within DU105) and non-real-time functions to be performed in a centralized manner (e.g., within CU108).

[0040] It should be understood that the distribution of functions between core network operations and access node operations may differ from, or even be absent from, the distribution of functions in LTE. Several other technological advancements that may be used, including big data and all-IP, may change how networks are built and managed. 5G (or New Radio, NR) networks may be designed to support multiple layers in which MEC servers may be placed between the core and access nodes. It should be recognized that MEC may also be applied to 4G networks.

[0041] 5G may utilize non-terrestrial communications, such as satellite communications, to increase or complement the coverage of 5G services by providing backhauling. Possible use cases may include providing service continuity for machine-to-machine (M2M) or Internet of Things (IoT) devices or for passengers on vehicles, or ensuring service availability for critical communications and future rail / maritime / air communications. Satellite communications may utilize geostationary earth orbit (GEO) satellite systems, or similarly, low earth orbit (LEO) satellite systems, particularly mega-constellations (systems with hundreds of (nano) satellites deployed). A given satellite 106 within a mega-constellation may cover several satellite-enabled network entities that create on-ground cells. On-ground cells may be created through on-ground relay nodes or by access nodes 104 located on-ground or within satellites.

[0042] 6G networks are expected to adopt flexible, decentralized and / or distributed computing systems and architectures, as well as ubiquitous computing, with local spectrum licensing, spectrum sharing, infrastructure sharing, and intelligent automated management supported by mobile edge computing, artificial intelligence, short packet communications, and blockchain technology. Key features of 6G may also include intelligent connected management and control capabilities, programmability, integrated sensing and communications, reduced energy footprint, reliable infrastructure, scalability, and affordability. In addition to these, 6G will also direct new use cases, such as the integration of location and sensing capabilities, toward a system definition for unifying user experiences across the physical and digital worlds.

[0043] It will be apparent to those skilled in the art that the system shown is only one example of a wireless access system, and in practice, the system may comprise multiple access nodes, a user device may have access to multiple wireless cells, and the system may comprise other devices such as physical layer relay nodes or other network elements. At least one of the access nodes may be a Home eNodeB or a Home gNodeB.

[0044] Furthermore, within the geographical area of ​​the wireless communication system, multiple different types of radio cells and multiple radio cells may be provided. Radio cells may be large cells (macrocells or umbrella cells) with a diameter of up to several tens of kilometers, or small cells such as microcells, femtocells, or picocells. The access nodes in Figure 1 may provide any type of these nodes. The cellular wireless system may be run as a multilayer network containing several types of radio cells. In a multilayer network, one access node may provide one or more radio cells of one type, and therefore multiple access nodes may be required to provide such a network structure.

[0045] To meet the need for improving the deployment and performance of communication systems, the concept of a "plug-and-play" access node may be introduced. A network that may use a "plug-and-play" access node may include a Home Node B gateway or HNB-GW (not shown in Figure 1) in addition to a Home eNodeB or Home gNodeB. An HNB-GW, which may be installed within the operator's network, may aggregate traffic from multiple Home eNodeBs or Home gNodeBs to return it to the core network.

[0046] Positioning techniques may be used to estimate the physical location of a user device. In this specification, a user device being positioned is referred to as a target UE or target user device. For example, the positioning technique used in NR may be based on at least one of the following: time difference of arrival (TDoA), time of arrival (TOA), time of departure (TOD), round trip time (RTT), angle of departure (AoD), angle of arrival (AoA), and / or carrier phase.

[0047] In Uu positioning (UL / DL positioning), multiple transmission and reception points (TRPs) within a known location may transmit and / or receive one or more positioning reference signals (PRS) to and from a target UE. In the uplink, a sounding reference signal (SRS) may be used as a positioning reference signal. For example, a multilateration technique may then be used to locate (position) the target UE with respect to the TRPs. One of these TRPs may be used as a positioning anchor, and the TDoA difference may be calculated with respect to this positioning anchor. The positioning anchor may be referred to herein as an anchor, anchor node, multilateration anchor, or reference point.

[0048] Sidelink (SL) positioning refers to a positioning approach in which a target UE utilizes a sidelink (i.e., a direct device-to-device link) to position itself either in an absolute way (in the case of absolute positioning) or in a relative way (in the case of relative positioning). In the case of UE-assisted positioning (in SL positioning and Uu positioning), the target UE may use the sidelink to obtain positioning measurements and report the measurements to a network entity such as a Location Management Function (LMF). Sidelink positioning may also be used to obtain distance information.

[0049] Distancing refers to determining the distance between two UEs and / or the direction of one UE from the other UE via a direct device connection.

[0050] Absolute positioning means estimating the location of a target UE in two- or three-dimensional geographic coordinates within a coordinate system (e.g., latitude, longitude, and / or altitude).

[0051] Relative positioning means estimating the location of a target UE relative to other network elements or other UEs.

[0052] SL positioning may also be based on the transmission of sidelink positioning reference signals (SL PRS) by multiple anchor UEs (anchor user devices), which are received and measured by the target UE to enable the target UE's location within the precise latency and accuracy requirements of the corresponding SL positioning session. Alternatively or additionally, the target UE may transmit SL PRS that are received and measured by the anchor UEs.

[0053] An anchor UE may be defined as a UE-supported positioning system for a target UE by, for example, transmitting and / or receiving a reference signal (e.g., SL PRS) for positioning via an SL interface. This may be similar to UL / DL-based positioning, where a gNB may serve as an anchor transmitting and / or receiving a reference signal to and from the target UE for positioning.

[0054] SL PRS refers to the reference signal transmitted through the SL for positioning purposes.

[0055] Furthermore, a positioning reference unit (PRU) may be used in a positioning session to improve the positioning accuracy for positioning a target UE. The PRU is a reference device at a known location that refines the location estimation of the target UE within that area and performs measurements used to generate correction data that may be used to improve the positioning accuracy for positioning the target UE. For example, a UE with a known location may be used as a PRU.

[0056] A PRU at a known location may perform positioning measurements, such as reference signal time difference (RSTD), reference signal received power (RSRP), and UE receive / transmit time difference measurements, and report these measurements to a location server such as an LMF. Furthermore, the PRU may transmit a UL SRS for positioning, enabling the TRP to measure and report UL positioning measurements (e.g., relative time of arrival, UL-AoA, gNB receive / transmit time difference, etc.) from the PRU at the known location. The PRU measurements may be compared by the location server to expected measurements at the known PRU location to determine correction data for other nearby target UEs. DL and / or UL location measurements for other target UEs may then be corrected based on the predetermined correction data.

[0057] The PRU may serve as a positioning anchor for the target UE, or it may simply provide correction data (e.g., to the LMF) to help position the target UE.

[0058] In other words, PRUs located at known locations may also act as reference target UEs, thereby comparing their calculated locations to those known locations. The comparison between known and estimated locations may yield correction data, which can be used for the location estimation process of other nearby target UEs, under the assumption that the same or similar accuracy determination effects apply to both the PRU locations and the locations of the other target UEs. The correction data may then be used to fine-tune the location estimation of the target UEs, thereby improving positioning accuracy.

[0059] The location of the target UE can be calculated, for example, in the LMF network (in the case of LMF-based positioning) or in the target UE's own network (in the case of UE-based positioning). Measurements for positioning can be performed on the UE side (e.g., in the case of DL or SL positioning) or on the network side (e.g., in the case of UL positioning).

[0060] Figure 2 shows an example in which one or more PRUs 202, 202A, and 202B are used to position a target UE 200. The PRUs may be configured to transmit a reference signal that is measured to position the target UE 200. The target UE 200 may further transmit a reference signal to position the target UE 200. One or more access nodes 204, 204A, and 204B may measure the reference signals received from PRUs 202, 202A, and 202B and from the target UE 200. In the case of side-link positioning, the target UE 200 may measure the reference signals received from PRUs 202, 202A, and 202B, and / or PRUs 202, 202A, and 202B may measure the reference signals received from the target UE 200 and / or from other UEs or PRUs. The parameters under measurement (measurement data) derived from the received reference signal may include, for example, the reference signal reception time, the reference signal time difference (RSTD), the reference signal arrival angle, and / or the RSRP. The measurement data may be reported to a network element acting as a Location Management Function (LMF) 212 configured to perform positioning based on the measurement data. The LMF 212 may estimate the location of the target UE 200 based on the received measurement data and the known locations of the PRUs measured by the reporting access node. For example, a location estimation function used in real-time kinematic positioning (RTK) applications of a global navigation satellite system may be used. As an example, if the measurements indicate that the signals received from one of the PRUs of the target UE 200 and PRU 202 are highly correlated, the location of the target UE 200 may be estimated to be closer to PRU 202 and further away from the other PRUs, 202A and 202B. The correction from a given PRU202A, 202B location may be calculated based on measurement data, for example, by using the difference between the measurement data associated with the target UE200 and the measurement data associated with the nearest PRU202.For example, multilateral measurements (multiple measurements of RSRP, RSTD, and / or other parameters) may indicate that the target UE200 is in a particular direction from the nearest PRU202, and that correction may be made in that direction.

[0061] The NR air interface may be enhanced with features that enable support for artificial intelligence (AI) and / or machine learning (ML) based algorithms for improved performance and / or reduced complexity and overhead. Some use cases for such AI / ML techniques may include (but are not limited to) improved channel state information (CSI) feedback (e.g., reduced overhead, improved accuracy, predictive), beam management (e.g., beam prediction in the time and / or spatial domains for reduced overhead and latency, improved beam selection accuracy), and improved positioning accuracy (e.g., in scenarios relating to positioning accuracy with heavy non-line-of-sight, reduced positioning reference signal transmission and measurement reporting overhead, limited availability of labeled data, and scenarios where the device has significant RF faults / defects affecting positioning measurements).

[0062] Using AI / ML techniques to improve positioning accuracy may involve training for positioning being performed in a central ML unit such as a Location Management Function (LMF) or a 5G Network Data Analytics Function (NWDAF). The NWDAF may perform data analytics to generate insights and take action to enhance the user experience, including positioning use cases.

[0063] Training of ML models in the central ML unit may be carried out, for example, according to the following process (described in steps 1-4 below). 1) A set of data acquisition devices may be deployed at a selected location. Alternatively, the data acquisition devices may be randomly selected from a given geographical area. For example, these data acquisition devices may be PRUs. Hereinafter, data acquisition devices will be referred to as PRUs for simplicity, but any other type of data acquisition device may be used instead of a PRU. 2) The PRU performs field positioning measurements and reports the measured values ​​to the central ML unit. 3) The central ML unit uses an emulation tool to generate (emulated) positioning measurements. It should be noted that step 3 may be performed instead of or in addition to step 2. 4) The central ML unit trains a general-purpose ML-based localization framework using the above positioning measurements (reported measurements and / or emulated measurements). In this specification, this general-purpose ML-based localization framework is referred to as GLoc.

[0064] The trained GLoc may then be deployed to network nodes that execute the ML process and / or algorithms. Such network nodes are referred to herein as hosts. Hosts may be of different types, and the type may be defined in relation to the host's radio frequency (RF) and / or baseband capability, form factor, or key performance indicators (KPIs). The hosts that execute the ML process may be, for example, target UEs, PRUs, and radio access networks (e.g., gNBs, TRPs, and / or location management components, LMCs) to improve positioning accuracy.

[0065] However, a problem with the General-Purpose ML-Based Localization Framework (GLoc) is that it does not address the specific RF limitations (also known as RF defects) of the deployed host type (e.g., handheld UE, roadside unit, or gNB). RF limitations / defects may depend on hardware limitations such as different antenna configurations and form factors, analog-to-digital conversion (ADC) resolution, and crystal oscillators. Various RF defects may introduce combinations of carrier frequency offset, sampling time offset, transmit / receive beam offset, clock offset and drift, and phase noise.

[0066] These RF defects can translate into further phase rotation and delay of the positioning signal by the RF chain, as observed in baseband receivers. As a result, positioning entities hosting the GLoc framework (e.g., UEs, TRPs, etc.) may experience certain RF base signal delays / rotations that are not taken into account within the GLoc and are inaccurately absorbed into positioning measurements. Such defects may differ for different host types. For example, a PRU, UE, or gNB hosting the GLoc may need to adapt the GLoc to its own RF-specific characteristics.

[0067] Therefore, currently, there may not be a one-size-fits-all GLoc framework that meets the high positioning accuracy requirements of NR Release 18, for example, at the centimeter level. Consequently, a given host type device may need to adapt the GLoc to its own characteristics, such as antenna form factor and configuration.

[0068] Several exemplary embodiments are described below, however, using the principles and terminology of 5G NR technology, without limiting the exemplary embodiments to 5G NR communication systems.

[0069] Some exemplary embodiments may address the above problem by providing a method for adapting a general-purpose GLoc framework for host-type specific ML positioning. In other words, some exemplary embodiments may be used to adapt a general-purpose machine learning model (e.g., trained using training data collected from different types of NR elements) to a machine learning model adapted to compensate for intrinsic errors specific to a given host type. Thus, some exemplary embodiments may use AI / ML techniques to provide improved positioning accuracy.

[0070] In the context of AI / ML, meta-learning refers to adapting a general-purpose model (e.g., one trained using features extracted from heterogeneous sources) to a specific type of entity and / or task. A sub-branch of meta-learning is transfer learning (TL). TL aims to adjust an already trained model to perform the same task but on different entity types.

[0071] Some exemplary embodiments may provide a TL framework for positioning, through which a general-purpose GLoc framework for positioning may be customized for a specific NR element host type. More specifically, before the GLoc framework is deployed on a large scale, the GLoc may be refined based on at least intrinsic features (e.g., RF constraints) of the NR element type (e.g., target UE, PRU, or gNB).

[0072] In some exemplary embodiments, a central ML unit (e.g., LMF) is enabled to select an NR head unit (NR-HU) as a representative of a given NR element type and, therefore, a given expected intrinsic strain range. The GLoc model may then be refined by or with the help of the NR-HU, thereby customizing the GLoc model to compensate for strains specific to its NR element type.

[0073] To implement model adaptation, the central ML unit may provide the NR-HU with a general-purpose GLoc framework, along with parameters (such as the capabilities of the device used to generate the GLoc and RF defects) that are expected to acquire such a framework and corresponding training procedures.

[0074] The generic GLoc may then be refined based on the individual types of NR-HUs (e.g., capability and RF defects). Finally, the adapted model may be reported to the central ML unit along with the reasons behind refining the process in that way. Based on such reasons, the central ML unit may iterate and further refine or validate the GLoc framework and provide the next refined version to the host entity.

[0075] As an example, a GLoc trained using UL SRS collected by TRP may be fitted to a static UE by using DL PRS as input, as observed in the static UE baseband. The refined model, called staticUE-GLoc, may be transferred back to an LMF, which then distributes it to other static UEs.

[0076] As another example, a GLoc trained using UL SRS collected by TRP may be fitted to a fast UE.

[0077] As another example, a GLoc trained in an outdoor environment may be adapted to an indoor environment to provide a more refined model called an indoor GLoc.

[0078] As another example, a GLoc trained on samples collected from urban scenarios may be adapted to suburban scenarios to provide a more refined model called a suburban GLoc.

[0079] Training in an NR-HU can be beneficial because the host (NR-HU) collects signals that are distorted similarly to other NR elements of the same type. TL by an NR-HU may be based on the fact that intrinsic signal distortion is specific to the positioning signals collected by the NR-HU, and that the NR-HU uses type-specific cost functions and type-specific model constraints (e.g., depth of the artificial or simulated neural network, available activation function, etc.) to refine the GLoc.

[0080] Alternatively or additionally, the central ML unit may request the NR-HU to collect timestamps, transfer its training data to the central ML unit, and specify its model constraints (if any) in order to refine the GLoc at the central ML unit.

[0081] Figure 3 shows a signal transmission diagram according to an exemplary embodiment. Two types of network nodes (type x and type y) are shown in Figure 3, but it should be noted that the number of types may differ from two. In other words, there may be one or more types of network nodes. Furthermore, the signal transmission procedure shown in Figure 3 may be extended and applied according to the actual number of types. A central ML unit (e.g., LMF) may determine the actual number of types depending on its ability to group various network nodes based on their RF characteristics.

[0082] Referring to Figure 3, in block 301, a central ML unit such as an LMF acquires a machine learning model for positioning, and the machine learning model is trained on first training data associated with one or more first network nodes. The machine learning model may be trained for positioning, or for similar tasks that may be used for positioning. The LMF may acquire the machine learning model by training the model in the LMF. Alternatively, the machine learning model may be trained in another entity from which the LMF may receive the machine learning model. The machine learning model may be referred to herein as GLoc. For example, the machine learning model may include an artificial neural network (ANN). An example of an artificial neural network is shown in Figure 10.

[0083] The first training data may include at least one of the following: reference signal measurement information measured at one or more first network nodes from one or more received positioning reference signals (e.g., DL PRS, UL SRS, and / or SL PRS); emulated reference signal measurement information; or simulated reference signal measurement information associated with one or more first network nodes. For example, the reference signal measurement information may include at least channel impulse response (CIR) measurements that may be simulated or measured at one or more first network nodes from one or more received positioning reference signals. Emulated reference signal measurement information may be obtained, for example, by using emulation tools such as ray tracing or digital twins.

[0084] One or more first network nodes may comprise one or more network nodes of different types. In one example, one or more first network nodes may comprise multiple network nodes of different types.

[0085] In block 302, the LMF defines error ranges for types x and y. The error range may indicate an internal timing error that occurred during measurement acquisition due to an RF defect or fault. For example, the transmit timing error may indicate the time delay from when the digital signal is generated in the baseband until the RF signal is transmitted from the transmitting antenna. The receive timing error may indicate the time delay from when the RF signal arrives at the receiving antenna until the signal is digitized and timestamped in the baseband. Error ranges may also be defined by the LMF to classify network nodes into types x and y. For example, a type x host may be associated with an error range -x of + / -a nanoseconds (e.g., a = 5 nanoseconds).

[0086] In block 303, the LMF selects a third network node of type x (indicated as type x NR-HU) which will perform transfer learning, i.e., update / refine the machine learning model (generic GLoc), to account for RF defects specific to type x. The type of the third network node may differ from the types of one or more of the first network nodes.

[0087] Therefore, a single type x NR-HU may be selected as a representative of all type x network nodes, and this single type x NR-HU may perform model adaptation. This limits the computational complexity and signal transmission overhead of transfer learning by avoiding a scheme in which each type x network node performs model adaptation independently.

[0088] The LMF may select a fourth network node of type y (indicated as type y NR-HU) which will update / refine the general-purpose GLoc to account for RF defects specific to type y. In this specification, type y may refer to a type different from type x.

[0089] In this specification, the term “network node” may mean, for example, a target user device, positioning reference unit, anchor user device, TRP, or access node (e.g., gNB) of a wireless access network.

[0090] In this specification, the term “type” may mean, for example, a vendor-specific user device, a vendor-specific access node (e.g., gNB), a TRP with specific RF characteristics, a user device with a specific number of receiving antennas, an industrial Internet of Things (IIoT) device, a low-power high-accuracy positioning (LPHAP) device, a RedCap device, a handheld user device, or a road-side unit (RSU). For example, type N may be a UE with N receiving antennas, type 1 may be a UE with one receiving antenna, type 2 may be a UE with two receiving antennas, and so on. A type may be defined in relation to both the target positioning accuracy and the inherent error range that a given network node of that type is expected to introduce.

[0091] In block 304, the LMF sends a request to a third network node (type x NR-HU) to update the machine learning model at the third network node based on the second training data (i.e., a request to assist with transfer learning). For example, the request may be sent in the information element of an LTE positioning protocol (LPP) request message. The third network node may accept or reject the request from the LMF based on the load status and / or hardware limitations of the third network node.

[0092] The request may include information about the configuration of the machine learning model to be updated. For example, such information may define the output and cost functions of the machine learning model (i.e., model capabilities), the type, size, and shape of the inputs to the machine learning model, as well as the architecture of the machine learning model.

[0093] As an unrestrictive example, the output of a machine learning model may be time-of-arrival (TOA) information used for positioning, and the cost function may be the mean squared error (MSE).

[0094] As an unrestricted example, the input to a machine learning model may be a certain number of strongest channel impulse response (CIR) complex gains.

[0095] As an unrestricted example, the architecture of a machine learning model may refer to a deep neural network (DNN) having N=10 hidden layers and an activation function of rectified linear unit (ReLU).

[0096] In block 305, the LMF sends a request to the fourth network node (type y NR-HU) to update the machine learning model at the fourth network node based on the third training data. For example, the request may be sent in the information element of the LPP request message. The fourth network node may accept or reject the request from the LMF based on the load status and / or hardware limitations of the fourth network node.

[0097] The request may include information about the configuration of the machine learning model to be updated. For example, such information may define the output and cost functions of the machine learning model (i.e., model capabilities), the type, size, and shape of the inputs to the machine learning model, as well as the architecture of the machine learning model.

[0098] In other words, in blocks 304 and 305, the LMF requires the configuration of different types of NR-HUs related to processing specific positioning requests.

[0099] In block 306, the third network node (type x NR-HU) sends a response message to the LMF to accept the request. For example, acceptance may be indicated in the information element of the LPP reply message, which may include yes, no flag, or conditional yes, where the model constraints are described. For example, the constraints may mean that type x NR-HU may support a maximum architecture different from the architecture configured for GLoc. For example, the reply may include "model_constraint=Model architecture:DNN with N= 4 hidden" to indicate, for example, that type x NR-HU supports a DNN with a maximum of 4 hidden layers, while the GLoc architecture may include a DNN with 10 hidden layers.

[0100] In block 307, the LMF transmits or forwards a machine learning model to the third network node (type x NR-HU) in response to the third network node accepting the request. This transmission may include at least one of the following: the structure of the machine learning model, one or more activation functions of the machine learning model, a set of weights for one layer of the machine learning model, a set of biases for one layer of the machine learning model, a cost function used to train the machine learning model, the input type and format of the machine learning model (i.e., how the inputs are arranged and what the inputs correspond to), and / or the output type and format of the machine learning model (e.g., a probability vector or a binary vector, vector length, etc.).

[0101] The input to the machine learning model may include at least one of the following: a received signal sample for one received beam for the total number of beams (some entries may be zero-padded if unavailable), reference signal received power (RSRP) for one positioning source and / or one beam, line-of-sight (LOS) indication or probability for one positioning source, etc. For example, the input may be a vector of CIRs of a certain length and entries arranged in decreasing order of magnitude.

[0102] For example, the output may be time of arrival (TOA) information represented as a real scalar.

[0103] The LMF may transmit information indicating a set of constraints for updating the machine learning model at a third network node. Thus, the LMF may parameterize the transfer learning procedure at least partially at the third network node. For example, the set of constraints may indicate updating the machine learning model using a reference signal from a selected pool of physical resource blocks (PRBs) for a given duration, and / or that certain conditions are met, or at least one of these. For example, the conditions may be met if the third network node considers itself to be static, uninterrupted, etc. Alternatively or additionally, the set of constraints may indicate, for example, freezing a portion of the machine learning model and updating the rest of the architecture in order to update the weights from layer L onwards.

[0104] The LMF may transmit information about a reference training procedure to update the machine learning model at a third network node. For example, the reference training procedure may correspond to using training data collected from multiple network nodes. Therefore, the information about the reference training procedure may include at least a set of parameters (e.g., capabilities and RF defects) used to construct the reference training procedure, and the set of parameters may be conceived by the third network node when refining the reference training procedure based on its own capabilities and defects.

[0105] In block 308, the fourth network node (type y NR-HU) sends a response message to the LMF to accept the request. For example, acceptance may be indicated in the information elements of the LPP reply message, which may include yes, no flag, or conditional yes, where the model constraints are described.

[0106] In block 309, in response to the fourth network node accepting the request, the LMF sends or transfers the machine learning model to the fourth network node (type y NR-HU). The LMF may send information indicating a set of constraints for updating the machine learning model at the fourth network node. The LMF may also send information about a reference training procedure for updating the machine learning model at the fourth network node, the information about the reference training procedure includes at least a set of parameters used to construct the reference training procedure.

[0107] The content of the transmission in block 309 may be the same as that described above for block 307.

[0108] In block 310, a third network node (type x NR-HU) obtains a first updated machine learning model by updating the machine learning model (i.e., the original GLoc) based on second training data. Updating may mean tuning, refining, or retraining the machine learning model based on second training data specific to the third network node of type x. The third network node may validate the first updated machine learning model. The first updated machine learning model obtained by the third network node may be referred to herein as type-x-GLoc.

[0109] The second training data may include reference signal measurement information measured at a third network node from one or more received positioning reference signals (e.g., DL PRS, UL SRS, and / or SL PRS), the third network node may be different from one or more first network nodes. For example, the reference signal measurement information may include at least channel impulse response (CIR) measurements measured at the third network node from one or more received positioning reference signals.

[0110] Updating a machine learning model may include the following steps 1-6. 1) A step in which the machine learning model is initialized using the model received from the LMF. Initialization may mean that the machine learning model is pruned or otherwise simplified according to the capabilities of the third network node. 2) A step of obtaining second training data, for example, a step of collecting or accessing it. 3) A step to clean and prepare the input by making the second training data conform to the format defined by LMF. 4) For example, by using a variation of the general cost function indicated by the LMF, select a cost function specific to the third network node, or a cost function from a list of cost functions selected by the LMF. 5) The step of selecting an activation function provided by the LMF to generate the output format defined by the LMF. 6) The step of performing updates / training under a set of constraints dictated by LMF.

[0111] In block 311, the third network node (type x NR-HU) sends a message to the LMF indicating the first updated machine learning model (type x GLoc) acquired by the third network node. The message may include the first updated machine learning model. Alternatively, the message may include an updated set of weights and biases associated with the first updated machine learning model. In other words, the refinement process may be reported to a provided reference process as a "delta" so that only the updated weights and biases can be reported (i.e., without reporting the entire model).

[0112] Messages may be sent based on whether the estimated performance improvement of the updated machine learning model exceeds a threshold. In other words, a refined model may be reported if, when tested, it produces a performance improvement greater than a given threshold (compared to the original GLoc). The threshold may be defined based on positioning accuracy and measurement granularity. This may reduce unnecessary reporting and therefore reduce network signal transmission.

[0113] In block 312, the fourth network node (type y NR-HU) obtains a second updated machine learning model by updating the machine learning model (i.e., the original GLoc) based on the third training data. The fourth network node may validate the second updated machine learning model. The second updated machine learning model obtained by the fourth network node may be called type-y-GLoc. The fourth network node may perform updates in the same manner as described above for block 310.

[0114] The third training data may include reference signal measurement information measured at a fourth network node from one or more received positioning reference signals (e.g., DL PRS, UL SRS, and / or SL PRS), the fourth network node may be different from one or more first network nodes and the third network node. For example, the reference signal measurement information may include at least channel impulse response (CIR) measurements measured at the fourth network node from one or more received positioning reference signals.

[0115] In block 313, the fourth network node (type y NR-HU) sends a message to the LMF indicating the second updated machine learning model (type y-GLoc) obtained by the fourth network node. The message may include the second updated machine learning model. Alternatively, the message may include an updated set of weights and biases associated with the second updated machine learning model.

[0116] In block 314, the LMF may validate or modify the first and second updated machine learning models. For example, prior to mass deployment, the LMF may cross-validate the updated models to ensure they remain robust and perform within target key performance indicators (KPIs). For example, the LMF may use stored test data to check that the updated models meet given KPI targets.

[0117] In block 315, the LMF transmits or distributes the first updated machine learning model (type-x-GLoc) to one or more second network nodes of type x. In other words, the LMF distributes the first updated machine learning model to a third network node (type x NR-HU) and other entities of the same type. One or more second network nodes may be referred to herein as type x units. The difference between a type x NR-HU and a type x unit is that a type x NR-HU has the ability to train the updated machine learning model, i.e., has the ability and computational resources to collect and label second training data, and is therefore selected to produce an updated machine learning model that works well with all type x units.

[0118] A given second network node (type x unit) may be configured to use a first updated machine learning model (type-x-GLoc) to position a target UE. For example, if type x unit is a gNB or anchor UE, type x unit may measure a reference signal received from the target UE, for example, to obtain CIR measurements. Alternatively, if type x unit is the target UE, type x unit may measure a reference signal received from the gNB or anchor UE, for example, to obtain CIR measurements. Type x unit may then provide these measurements as input to type-x-GLoc. The output of type-x-GLoc may be a location estimate of the target UE. Alternatively, the output of type-x-GLoc may be some other useful positioning-related information or intermediate features, such as time of arrival (TOA) and / or angle of arrival (AOA) for (possible) in-line-of-sight (LOS) and / or strong beyond-line-of-sight (NLOS) paths.

[0119] A type-x-GLoc may be fed the same input information and produce the same output type as the original GLoc. The difference lies in the accuracy of these models, and the type-x-GLoc model may provide higher accuracy for a particular type of device (i.e., for type x units) compared to the original GLoc, because the type-x-GLoc model is refined based on specific RF defects of this device type.

[0120] It should be noted that the LMF may send the first updated machine learning model to other types of network nodes, for example, one or more first network nodes.

[0121] In block 316, the LMF sends or distributes type-y-GLoc to one or more fifth network nodes of type y. In other words, the LMF distributes the second updated machine learning model to other entities of the same type as the fourth network node (type y NR-HU).

[0122] It should be noted that the LMF may send the second updated machine learning model to other types of network nodes, for example, one or more first network nodes.

[0123] In this specification, terms such as "first network node," "second network node," etc., are used to distinguish network nodes and do not necessarily refer to specific identifiers of network nodes.

[0124] Figure 4 shows a signal transmission diagram according to another exemplary embodiment. In this exemplary embodiment, a selected NR-HU collects its training data and transfers it back to the LMF. The LMF may then aggregate the data from multiple network nodes of the same type, store it in memory, and generate an updated machine learning model. It should be noted that the LMF may use the collected training data and combine it across different network node types used when updating a general-purpose GLoc model.

[0125] Two types of network nodes (type x and type y) are shown in Figure 4, but it should be noted that the number of types may differ from two. In other words, there may be one or more types of network nodes. Furthermore, the signal transmission procedure shown in Figure 4 may be extended and applied according to the actual number of types. A central ML unit (e.g., LMF) may determine the actual number of types depending on its ability to group the various network nodes based on their RF characteristics.

[0126] Referring to Figure 4, in block 401, a central ML unit such as an LMF acquires a machine learning model for positioning, and the machine learning model is trained on first training data associated with one or more first network nodes. The machine learning model may be trained for positioning, or for similar tasks that may be used for positioning. The LMF may acquire the machine learning model by training the model in the LMF. Alternatively, the machine learning model may be trained in another entity from which the LMF may receive the machine learning model. The machine learning model may be referred to herein as GLoc. For example, the machine learning model may include an artificial neural network (ANN). An example of an artificial neural network is shown in Figure 10.

[0127] The first training data may include at least one of the following: reference signal measurement information measured at one or more first network nodes from one or more received positioning reference signals (e.g., DL PRS, UL SRS, and / or SL PRS); emulated reference signal measurement information; or simulated reference signal measurement information associated with one or more first network nodes. For example, the reference signal measurement information may include at least channel impulse response (CIR) measurements that may be simulated or measured at one or more first network nodes from one or more received positioning reference signals. The emulated reference signal measurement information may be obtained, for example, by using emulation tools such as ray tracing or digital twins.

[0128] One or more first network nodes may comprise one or more network nodes of different types. In one example, one or more first network nodes may comprise multiple network nodes of different types.

[0129] In block 402, the LMF defines error ranges for types x and y. The error range may indicate an internal timing error that occurred during measurement acquisition due to an RF defect or fault. For example, the transmit timing error may indicate the time delay from when the digital signal is generated in the baseband until the RF signal is transmitted from the transmitting antenna. The receive timing error may indicate the time delay from when the RF signal arrives at the receiving antenna until the signal is digitized and timestamped in the baseband. Error ranges may also be defined by the LMF to classify network nodes into types x and y. For example, a type x host may be associated with an error range -x of + / -a nanoseconds (e.g., a = 5 nanoseconds).

[0130] In block 403, the LMF selects one or more third network nodes of type x (indicated as type x NR-HU) from which it will request training data to update / refine its machine learning model (generic GLoc) in order to account for RF defects specific to type x. The type of one or more third network nodes may differ from the type of one or more first network nodes.

[0131] The LMF may select one or more fourth network nodes of type y (indicated as type y NR-HU) from which the LMF will request training data to update / refine a machine learning model (generic GLoc) in order to account for RF defects specific to type y. In this specification, type y may refer to a type different from type x.

[0132] In this specification, the term “network node” may mean, for example, a target user device, positioning reference unit, anchor UE, TRP, or access node (e.g., gNB) of a wireless access network.

[0133] In this specification, the term “type” may mean, for example, a vendor-specific user device, a vendor-specific access node (e.g., gNB), a TRP with specific RF characteristics, a user device with a specific number of receiving antennas, an industrial Internet of Things (IIoT) device, a low-power high-precision positioning (LPHAP) device, a capacity reduction (RedCap) device, a handheld user device, or a roadside unit (RSU). A type may be defined in relation to both the target positioning accuracy and the inherent error range that a given network node of that type is expected to introduce.

[0134] In block 404, the LMF sends a request to one or more third network nodes (type x NR-HU) to provide the LMF with second training data from one or more third network nodes in order to update the machine learning model in the LMF. For example, the request may be sent in the information element of the LPP request message. A given third network node may accept or reject the request from the LMF based on the load status and / or hardware limitations of the third network node.

[0135] In block 405, the LMF sends a request to one or more fourth network nodes (type y NR-HU) to provide the LMF with third training data from one or more of the fourth network nodes in order to update the machine learning model in the LMF. For example, the request may be sent in the information element of the LPP request message. A given fourth network node may accept or reject the request from the LMF based on the load status and / or hardware limitations of the fourth network node.

[0136] In block 406, one or more third network nodes (type x NR-HU) send a response message to the LMF to accept the request. For example, acceptance may be indicated in the information element of the LPP reply message, which may include a yes or no flag.

[0137] In block 407, one or more fourth network nodes (type y NR-HU) send a response message to the LMF to accept the request. For example, acceptance may be indicated in the information element of the LPP reply message, which may include a yes or no flag.

[0138] In block 408, one or more third network nodes (type x NR-HU) acquire second training data. For example, one or more third network nodes may acquire second training data by performing measurements, such as CIR measurements, on a received positioning reference signal such as DL PRS, UL SRS, or SL PRS. Alternatively, one or more third network nodes may acquire second training data by retrieving stored measurements from memory. One or more third network nodes may clean the second training data before transmitting it to the LMF.

[0139] In block 409, one or more third network nodes (type x NR-HU) send second training data to the LMF. The LMF may store the received second training data in its memory. One or more third network nodes may send information to the LMF indicating a first set of constraints for updating the machine learning model. For example, the first set of constraints may include the artificial neural network, the available activation functions, and the maximum supported depth (i.e., the number of layers).

[0140] In block 410, one or more fourth network nodes (type y NR-HU) acquire third training data. For example, one or more fourth network nodes may acquire third training data by performing measurements such as CIR measurements on a received positioning reference signal such as DL PRS, UL SRS, or SL PRS. Alternatively, one or more fourth network nodes may acquire third training data by retrieving stored measurements from memory. One or more fourth network nodes may clean the third training data before transmitting it to the LMF.

[0141] In block 411, one or more fourth network nodes (type y NR-HU) send third training data to the LMF. The LMF may store the received third training data in its memory. One or more fourth network nodes may send information to the LMF indicating a second set of constraints for updating the machine learning model. For example, the second set of constraints may include the maximum supported depth (i.e., number of layers) of the artificial neural network, available activation functions, etc.

[0142] In block 412, the LMF obtains a first updated machine learning model (type-x-GLoc) by updating the machine learning model (original GLoc) at least partially based on a second set of training data received from one or more third network nodes.

[0143] The update may include one or more of the following steps 1-6. 1) A step to initialize the machine learning model using the original GLoc. 2) The step of accessing the second set of training data. 3) A step to clean and prepare the second training data to match the GLoc input format. 4) The step of selecting a cost function specific to one or more third network nodes, for example, by using a variation of GLoc's general cost function. 5) The step of selecting an activation function to generate the GLoc output format. 6) A step of performing updates / training under a set of constraints directed by one or more third network nodes.

[0144] The LMF may obtain a second updated machine learning model (type-y-GLoc) by updating the machine learning model (original GLoc) at least partially based on a third set of training data received from one or more fourth network nodes.

[0145] In block 413, the LMF transmits or distributes the first updated machine learning model (type-x-GLoc) to one or more second network nodes of type x. In other words, the LMF distributes the first updated machine learning model to one or more third network nodes (type x NR-HUs) and other entities of the same type. One or more second network nodes may be referred to herein as type x units. The difference between a type x NR-HU and a type x unit is that a type x NR-HU has the ability to train the updated machine learning model, i.e., the ability and computational resources to collect and label second training data, and is therefore selected to produce an updated machine learning model that works well with all type x units.

[0146] A given second network node (type x unit) may be configured to use a first updated machine learning model (type-x-GLoc) to position a target UE. For example, if type x unit is a gNB or anchor UE, type x unit may measure a reference signal received from the target UE, for example, to obtain CIR measurements. Alternatively, if type x unit is the target UE, type x unit may measure a reference signal received from the gNB or anchor UE, for example, to obtain CIR measurements. Type x unit may then provide these measurements as input to type-x-GLoc. The output of type-x-GLoc may be a location estimate of the target UE. Alternatively, the output of type-x-GLoc may be some other useful positioning-related information or intermediate features, such as time of arrival (TOA) and / or angle of arrival (AOA) for (possible) in-line-of-sight (LOS) and / or strong beyond-line-of-sight (NLOS) paths.

[0147] A type-x-GLoc may be fed the same input information and produce the same output type as the original GLoc. The difference lies in the accuracy of these models, with the type-x-GLoc model providing higher accuracy for a particular type of device (i.e., for type x devices) compared to the original GLoc, because the type-x-GLoc model is refined based on specific RF defects of this device type.

[0148] It should be noted that the LMF may send the first updated machine learning model to other types of network nodes, for example, one or more first network nodes.

[0149] In block 414, the LMF transmits or distributes type-y-GLoc to one or more fifth network nodes of type y. In other words, the LMF distributes the second updated machine learning model to other entities of the same type as the fourth network node (type y NR-HU). One or more fifth network nodes may be referred to herein as type y units.

[0150] It should be noted that the LMF may send the second updated machine learning model to other types of network nodes, for example, one or more first network nodes.

[0151] Figure 5 shows a flowchart illustrating an exemplary embodiment of a method implemented by a device that includes, or is equipped with, a Location Management Function (LMF) or a Network Data Analytics Function (NWDAF). For example, the device may correspond to LMF112 in Figure 1 or LMF212 in Figure 2.

[0152] Referring to Figure 5, in block 501, a machine learning model for positioning is acquired, and the machine learning model is trained on first training data associated with one or more first network nodes.

[0153] In block 502, information is transmitted that includes at least one of the following: a machine learning model, a request to update the machine learning model based on second training data, or a request to provide second training data in order to update the machine learning model in the device.

[0154] In block 503, at least one of the following is received: a message indicating an updated machine learning model, or second training data.

[0155] In block 504, the updated machine learning model is sent to at least one or more second network nodes.

[0156] Figure 6 shows a flowchart of an exemplary embodiment of a method implemented by a device such as a network node or a device comprising or included therein. A network node may refer to, for example, a user device, positioning reference unit, or access node of a wireless access network. For example, a network node may correspond to UE100, UE102, or access node 104 in Figure 1, or any of UE200, PRU202, 202A, 202B, or any of access nodes 204, 204A, 204B in Figure 2.

[0157] Referring to Figure 6, in block 601, information is received that includes at least one of a machine learning model for positioning, a request to update the machine learning model in the device based on second training data, or a request to provide second training data to update the machine learning model, and the machine learning model is trained based on first training data associated with one or more first network nodes.

[0158] In block 602, the second set of training data is acquired.

[0159] In block 603, at least one of the following is sent: a message indicating the updated machine learning model, or second training data.

[0160] As used herein, "at least one of the following: <list of two or more elements>" ) and at least one of The phrases ")" and similar expressions where a list of two or more elements is joined by "and" or "or" mean at least any one element of an element, or at least any two or more elements of an element, or at least all elements.

[0161] The blocks, associated functions, and information exchanges (messages) described above in Figures 3-6 are not in absolute chronological order, and some of them may be performed simultaneously or in an order different from the order described. Other functions may be performed similarly between or within them, other information may be sent, and / or other rules may apply. Some of the information in a block, part of a block, or one or more pieces of information may be omitted or replaced by the corresponding block, part of a block, or one or more pieces of information.

[0162] Figure 7 shows an example of a device 700 having means for implementing any method of the methods in Figures 3-6 or any other exemplary embodiment described above. For example, the device 700 may be a device that includes or comprises a user device. The user device may correspond to any user device 100, 102 in Figure 1 or any PRU 200 in Figure 2 or any PRU 202, 202A, 202B in Figure 2. The user device may also be called a subscriber unit, mobile station, remote terminal, access terminal, user terminal, terminal device, user equipment (UE), target UE, target user device, anchor UE, positioning reference unit (PRU), NR-HU, host, or network node.

[0163] The device 700 comprises at least one processor 710. The at least one processor 710 interprets instructions (e.g., computer program instructions) and processes data. The at least one processor 710 may comprise one or more programmable processors. The at least one processor 710 may comprise programmable hardware having embedded firmware, and may optionally or additionally comprise one or more application-specific integrated circuits (ASICs).

[0164] At least one processor 710 is coupled to at least one memory 720. At least one processor is configured to write to / read data from at least one memory 720. At least one memory 720 may comprise one or more memory units. Memory units may be volatile or non-volatile. It should be noted that there may be one or more memory units of non-volatile memory and one or more memory units of volatile memory, or alternatively, one or more memory units of non-volatile memory, or alternatively, one or more memory units of volatile memory. Volatile memory may be, for example, random access memory (RAM), dynamic random access memory (DRAM), or synchronous dynamic random access memory (SDRAM). Non-volatile memory may be, for example, read-only memory (ROM), programmable read-only memory (PROM), electrically erasable programmable read-only memory (EEPROM), flash memory, optical storage, or magnetic storage. In general, memory may be called non-temporary computer-readable medium. The term “non-transitory” as used herein refers to limitations of the medium itself (i.e., tangible and not signaling) as opposed to limitations relating to data storage persistence (e.g., RAM vs. ROM). At least one memory 720 stores computer-readable instructions, which are executed by at least one processor 710 to implement one or more of the exemplary embodiments described above. For example, non-volatile memory stores computer-readable instructions, and at least one processor 710 executes the instructions using volatile memory for temporary storage of data and / or instructions. Computer-readable instructions may also be called computer program code.

[0165] Computer-readable instructions may be pre-stored in at least one memory 720, or, alternatively or additionally, computer-readable instructions may be received by the device via electromagnetic carrier signals and / or copied from a physical entity such as a computer program product. Execution of computer-readable instructions by at least one processor 710 causes the device 700 to implement one or more of the exemplary embodiments described above. That is, at least one processor and at least one memory storing instructions may provide or bring about the performance of any of the methods and / or blocks described above.

[0166] In the context of this document, “memory” or “computer-readable media” or “computer-readable medium” may be any non-temporary medium or medium or means that can store, communicate, propagate or transport instructions for use by or in connection with an instruction execution system, apparatus, or device such as a computer. The term “non-temporary” is, as used herein, a limitation of the medium itself (i.e., tangible and not signaling) in contrast to a limitation relating to the persistence of data storage (e.g., RAM vs. ROM).

[0167] The device 700 may further include, or be connected to, an input unit 730. The input unit 730 may include one or more interfaces for receiving input. One or more interfaces may include, for example, one or more temperature, motion, and / or orientation sensors, one or more cameras, one or more accelerometers, one or more microphones, one or more buttons, and / or one or more touch detection units. Furthermore, the input unit 730 may include interfaces to which external devices may be connected.

[0168] The device 700 may include an output unit 740. The output unit may include, or be connected to, one or more displays capable of rendering visual content, such as a light-emitting diode (LED) display, a liquid crystal display (LCD), and / or a liquid crystal on silicon (LCoS) display. The output unit 740 may further include one or more audio outputs. One or more audio outputs may be, for example, loudspeakers.

[0169] The device 700 further comprises a connectivity unit 750, which enables wireless connectivity to one or more external devices. The connectivity unit 750 comprises at least one transmitter and at least one receiver, which may be integrated into the device 700 or to which the device 700 may be connected. The at least one transmitter comprises at least one transmitting antenna, and the at least one receiver comprises at least one receiving antenna. The connectivity unit 750 may comprise an integrated circuit or set of integrated circuits that provide wireless communication capabilities for the device 700. Alternatively, the wireless connectivity may be a hardwired application-specific integrated circuit (ASIC). The connectivity unit 750 may provide means for implementing at least some blocks of the blocks of one or more exemplary embodiments described above. The connectivity unit 750 may include one or more components controlled by a corresponding control unit, such as a power amplifier, a digital front end (DFE), an analog-to-digital converter (ADC), a digital-to-analog converter (DAC), a frequency converter, a (de)modulator, and / or an encoder / decoder circuit.

[0170] It should be noted that the device 700 may further comprise various components not shown in Figure 7. These components may be hardware components and / or software components.

[0171] Figure 8 shows an example of a device 800 having means for implementing any method of the methods in Figures 3-6 or any other exemplary embodiment described above. For example, device 800 may be an access node of a radio access network, or a device comprising or including such an access node. The access node may correspond to any access node such as access node 104 in Figure 1 or access nodes 204, 204A, 204B in Figure 2. Device 800 may be referred to, for example, a network node, a radio access network (RAN) node, a next-generation radio access network (NG-RAN) node, node B, eNB, gNB, base transceiver base station (BTS), base station, NR base station, 5G base station, access point (AP), relay node, repeater, integrated access and backhaul (IAB) node, IAB donor node, distributed unit (DU), central unit (CU), baseband unit (BBU), radio unit (RU), radio head, remote radio head (RRH), or transmit and receive point (TRP).

[0172] The device 800 may, for example, include a circuit unit or chipset applicable to implement one or more of the exemplary embodiments described above. The device 800 may also be an electronic device comprising one or more electronic circuit units. The device 800 may include a communication control circuit unit 810, such as at least one processor, and at least one memory 820 for storing instructions, which, when executed by at least one processor, cause the device 800 to implement one or more of the exemplary embodiments described above. Such instructions 822 may include, for example, computer program code (software), and at least one memory and computer program code (software) are configured to cause the device 800 to implement one or more of the exemplary embodiments described above using at least one processor. In this specification, computer program code may then refer to an instruction, which, when executed by at least one processor, causes the device 800 to implement one or more of the exemplary embodiments described above. In other words, at least one processor and at least one memory for storing instructions may provide or bring about the performance of any of the methods and / or blocks described above.

[0173] The processor is coupled to memory 820. The processor is configured to write to and read data from memory 820. Memory 820 may comprise one or more memory units. Memory units may be volatile or non-volatile. It should be noted that there may be one or more units of non-volatile memory and one or more units of volatile memory, or alternatively, one or more units of non-volatile memory, or alternatively, one or more units of volatile memory. Volatile memory may be, for example, random access memory (RAM), dynamic random access memory (DRAM), or synchronous dynamic random access memory (SDRAM). Non-volatile memory may be, for example, read-only memory (ROM), programmable read-only memory (PROM), electrically erasable programmable read-only memory (EEPROM), flash memory, optical storage, or magnetic storage. In general, memory may be referred to as non-transient computer-readable medium. The term “non-transient” as used herein is a limitation of the medium itself (i.e., tangible and not signaling) in contrast to a limitation on data storage persistence (e.g., RAM vs. ROM). Memory 820 stores computer-readable instructions that are executed by the processor. For example, non-volatile memory stores computer-readable instructions, and the processor uses volatile memory for temporary storage of data and / or instructions to execute those instructions.

[0174] Computer-readable instructions may be pre-stored in memory 820, or, alternatively or additionally, may be received by the device via electromagnetic carrier signals and / or copied from a physical entity such as a computer program product. Execution of a computer-readable instruction causes the device 800 to perform one or more of the functions described above.

[0175] The memory 820 may be implemented using any suitable data storage technology, such as semiconductor-based memory devices, flash memory, magnetic memory devices and systems, optical memory devices and systems, fixed memory, and / or removable memory. The memory may include a configuration database for storing configuration data. For example, the configuration database may store a current list of adjacent cells and, in some exemplary embodiments, the structure of frames used within the detected adjacent cells.

[0176] The device 800 may further include a communication interface 830 comprising hardware and / or software for achieving communication connectivity according to one or more communication protocols. The communication interface 830 comprises at least one transmitter (Tx) and at least one receiver (Rx), which may be integrated into the device 800 or to which the device 800 may be connected. The communication interface 830 may provide means for implementing some blocks of the blocks of one or more exemplary embodiments described above. The communication interface 830 may comprise one or more components, such as a power amplifier, a digital front end (DFE), an analog-to-digital converter (ADC), a digital-to-analog converter (DAC), a frequency converter, a (de)modulator, and / or an encoder / decoder circuit, which are controlled by a corresponding control unit.

[0177] The communication interface 830 provides a device having wireless communication capabilities for communicating in a cellular communication system. The communication interface may, for example, provide a wireless interface to one or more user devices. The device 800 may further include other interfaces for a core network such as a network coordinator device or AMF and / or for access nodes of the cellular communication system.

[0178] The device 800 may further include a scheduler 840 configured to allocate wireless resources. The scheduler 840 may be configured together with the communication control circuit unit 810, or the scheduler 840 may be configured separately.

[0179] It should be noted that the device 800 may further include various components not shown in Figure 8. These components may be hardware components and / or software components.

[0180] Figure 9 shows an example of a device 900 having means for implementing any method of the methods in Figures 3-6 or any other exemplary embodiments described above. For example, the device 900 may be a device such as a central ML unit, or a device comprising or including a central ML unit. The central ML unit may be called, for example, a Location Management Function (LMF), a Location Server, or a Network Data Analytics Function (NWDAF). For example, the central ML unit may correspond to LMF112 in Figure 1 or LMF212 in Figure 2.

[0181] The device 900 may, for example, include a circuit unit or chipset applicable to implement one or more of the exemplary embodiments described above. The device 900 may also be an electronic device comprising one or more electronic circuit units. The device 900 may include a communication control circuit unit 910, such as at least one processor, and at least one memory 920 for storing instructions, which, when executed by at least one processor, cause the device 900 to implement one or more of the exemplary embodiments described above. Such instructions 922 may include, for example, computer program code (software), and at least one memory and computer program code (software) are configured to cause the device 900 to implement one or more of the exemplary embodiments described above using at least one processor. In this specification, computer program code may then refer to an instruction, which, when executed by at least one processor, causes the device 900 to implement one or more of the exemplary embodiments described above. In other words, at least one processor and at least one memory for storing instructions may provide or bring about the performance of any of the methods and / or blocks described above.

[0182] The processor is coupled to memory 920. The processor is configured to write to and read data from memory 920. Memory 920 may comprise one or more memory units. Memory units may be volatile or non-volatile. It should be noted that there may be one or more units of non-volatile memory and one or more units of volatile memory, or alternatively, one or more units of non-volatile memory, or alternatively, one or more units of volatile memory. Volatile memory may be, for example, random access memory (RAM), dynamic random access memory (DRAM), or synchronous dynamic random access memory (SDRAM). Non-volatile memory may be, for example, read-only memory (ROM), programmable read-only memory (PROM), electrically erasable programmable read-only memory (EEPROM), flash memory, optical storage, or magnetic storage. In general, memory may be referred to as non-transient computer-readable medium. The term “non-transient” as used herein is a limitation of the medium itself (i.e., tangible and not signaling) in contrast to a limitation on data storage persistence (e.g., RAM vs. ROM). Memory 920 stores computer-readable instructions that are executed by the processor. For example, non-volatile memory stores computer-readable instructions, and the processor uses volatile memory for temporary storage of data and / or instructions to execute those instructions.

[0183] Computer-readable instructions may be pre-stored in memory 920, or, alternatively or additionally, may be received by the device via electromagnetic carrier signals and / or copied from a physical entity such as a computer program product. Execution of a computer-readable instruction causes the device 900 to perform one or more of the functions described above.

[0184] The memory 920 may be implemented using any suitable data storage technology, such as semiconductor-based memory devices, flash memory, magnetic memory devices and systems, optical memory devices and systems, fixed memory, and / or removable memory. The memory may include a configuration database for storing configuration data. For example, the configuration database may store a current list of adjacent cells and, in some exemplary embodiments, the structure of frames used within the detected adjacent cells.

[0185] The device 900 may further include a communication interface 930 comprising hardware and / or software for achieving communication connectivity according to one or more communication protocols. The communication interface 930 comprises at least one transmitter (Tx) and at least one receiver (Rx), which may be integrated into the device 900 or to which the device 900 may be connected. The communication interface 930 may provide means for implementing some blocks of the blocks of one or more exemplary embodiments described above. The communication interface 930 may comprise one or more components, such as a power amplifier, a digital front end (DFE), an analog-to-digital converter (ADC), a digital-to-analog converter (DAC), a frequency converter, a (de)modulator, and / or an encoder / decoder circuit, which are controlled by a corresponding control unit.

[0186] The communication interface 930 provides a device having wireless communication capabilities for communicating in a cellular communication system. The communication interface may, for example, provide a wireless interface to one or more user devices. The device 900 may further include another interface for a core network such as a network coordinator device or AMF and / or for access nodes of the cellular communication system.

[0187] It should be noted that the device 900 may further comprise various components not shown in Figure 9. These components may be hardware components and / or software components.

[0188] As used in this application, the term “circuitry” may mean one or more or all of the following: a) a hardware-only mode of execution of a circuit (such as a mode of execution of only analog and / or digital circuitry); b) a combination of hardware circuitry and software, for example, (where applicable), i) a combination of analog hardware circuitry and / or digital hardware circuitry having software / firmware; and ii) any part of a hardware processor having software (including digital signal processors, software, and memory that work together to enable a device such as a mobile phone to perform various functions); and c) hardware circuitry and / or processors such as a microprocessor or a part of a microprocessor that require software (e.g., firmware) for operation, but the software may not be present when not required for operation.

[0189] This definition of "circuit section" applies to all use of the term in this application, including in any claim. As a further example, as used in this application, the term "circuit section" also covers not only a hardware circuit or processor (or more processors), but also a given portion of a hardware circuit or processor and / or its associated software and / or firmware execution mode. The term "circuit section" also covers, for example, a baseband integrated circuit, or a processor integrated circuit for a mobile device, or a similar integrated circuit in a server, cellular network device, or other computing or network device, as applicable to a particular claim element.

[0190] Figure 10 shows an example of an artificial neural network 1030 having one hidden layer 1002, and Figure 11 shows an example of a computing node 1004. However, it should be noted that the artificial neural network 1030 may have two or more hidden layers 1002.

[0191] An artificial neural network (ANN) contains a set of rules designed to perform tasks such as regression, classification, clustering, and pattern recognition. An ANN may achieve such objectives through a learning / training procedure, where the objective is illustrated by various examples of input data along with the desired output. Thus, the ANN learns to identify the appropriate output for any given input within a training data manifold. Learning / training using labels is called supervised learning, while learning without labels is called unsupervised learning. Deep learning may require large amounts of input data.

[0192] Deep learning (also known as deep structured learning or hierarchical learning) is part of a broader family of layer-based machine learning models used in artificial neural networks. A deep neural network (DNN) 1030 is an artificial neural network with multiple hidden layers 1002 between an input layer 1000 and an output layer 1014. Training a DNN allows it to find the correct mathematical operations to translate an input into an appropriate output, even when the relationships are highly nonlinear and / or complex.

[0193] A given hidden layer 1002 comprises nodes 1004, 1006, 1008, 1010, and 1012 where calculations are performed. As shown in Figure 11, a given node 1004 combines input data 1000 with a set of coefficients or weights 1100 that amplify or attenuate its input 1000, thereby assigning significance to the input 1000 with respect to the task the algorithm attempts to learn. The input weight product is added together, and the sum, through the activation function 1104, determines whether and how far the signal should proceed through the neural network 1030 to influence the final outcome, such as the act of classification. In the process, the neural network learns to recognize the correlation between specific relevant features and the optimal outcome.

[0194] In the case of classification, the output of DNN1030 may be considered as a probability of a specific outcome. In this case, the number of layers 1002 may vary proportionally to the number of input data 1000 used. However, when the number of input data 1000 is large, the accuracy of the outcome 1014 is reliable. On the other hand, when there are a small number of layers 1002, the computation may take less time and therefore reduce latency. However, this is strongly dependent on the specific DNN architecture and / or available computing resources.

[0195] The initial weights 1100 of the model can be set in various alternative ways. During the training phase, the weights may be adapted to improve the accuracy of the process based on the analytical error in decision-making. Training the model is essentially a trial-and-error activity. In principle, given nodes 1004, 1006, 1008, 1010, and 1012 of the neural network 1030 make decisions (inputs). * The node assigns weights and then compares this decision to the collected data to find the difference relative to the collected data. In other words, the node determines the error on which the weights 1100 are adjusted. Thus, training the model can be thought of as a corrective feedback loop.

[0196] For example, a neural network model may be trained using a stochastic gradient descent optimization algorithm, for which the gradient is calculated using a backpropagation algorithm. The gradient descent algorithm attempts to change the weights 1100 so that the next evaluation decreases the error, and the optimization algorithm means navigating to lower the gradient (or slope) of the error. It is also possible to use any other suitable optimization algorithm if it provides sufficiently accurate weights 1100. As a result, the trained parameters of the neural network 1030 may include the weights 1100.

[0197] In the context of optimization algorithms, the function used to evaluate candidate solutions (i.e., sets of weights) is called the objective function. For neural networks, where the target is to minimize the error, the objective function may also be called the cost function or loss function. Any suitable method used when tuning the weights can be used as the loss function. Some examples of loss functions include mean squared error (MSE), maximum likelihood estimation (MLE), and cross-entropy.

[0198] For node 1004, the activation function 1104 defines the output 1014 of node 1004, given an input or set of inputs 1000. Node 1004 calculates a weighted sum of the inputs, possibly adding a bias, and then makes a decision as "activate" or "not activate" based on a decision threshold as a binary activation, or using the activation function 1104 which gives a nonlinear decision function. Any suitable activation function 1104, e.g., sigmoid, normalized linear unit (ReLU), normalized exponential function (softmax), softplus, tanh, etc., may be used. In deep learning, the activation function 1104 may be set at the layer level and applied to all neurons (nodes) within that layer. The output 1014 is then used as input for the next node, etc., until the desired solution to the original problem is found.

[0199] The techniques and methods described herein may be carried out by various means. For example, these techniques may be carried out in hardware (one or more devices), firmware (one or more devices), software (one or more modules), or a combination thereof. In the case of hardware implementation, the apparatus of the exemplary embodiment may be carried out in one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), graphics processing units (GPUs), processors, controllers, microcontrollers, microprocessors, other electronic units designed to perform the functions described herein, or a combination thereof. In the case of firmware or software, the implementation may be carried out through at least one chipset module (e.g., a procedure, a function, etc.) that performs the functions described herein. The software code may be stored in a memory unit and executed by a processor. The memory unit may be executed within or outside the processor. In the latter case, the memory unit can be communicatively coupled to the processor by various means, as is known in the art. Furthermore, the components of the system described herein may be rearranged and / or complemented by further components to facilitate the achievement of various embodiments, such as those described with respect to the components, and the components are not limited to the exact configuration shown in the given figures, as will be recognized by those skilled in the art.

[0200] As technology advances, it will be apparent to those skilled in the art that the concepts of the present invention may be implemented in a variety of ways. Embodiments are not limited to the exemplary embodiments described above, but may vary within the scope of the claims. Therefore, all words and expressions should be interpreted broadly and are intended to illustrate, rather than limit, the exemplary embodiments.

Claims

1. A device comprising at least one processor and at least one memory for storing instructions, wherein when an instruction is executed by the at least one processor, the device receives at least, The acquisition of a machine learning model for positioning, wherein the machine learning model is trained on first training data associated with one or more first network nodes. Transmitting information including at least one of the following: the machine learning model, a request to update the machine learning model based on second training data, or a request to provide the second training data for updating the machine learning model in the device; Receiving at least one of a message indicating an updated machine learning model, or the second training data, Transmit the updated machine learning model to at least one or more second network nodes. A device that performs an action.

2. Further selection of a third network node is made, and the information is transmitted to the third network node. The apparatus according to claim 1, wherein the type of the third network node is different from the type of the one or more first network nodes, and the type of the one or more second network nodes corresponds to the type of the third network node.

3. The apparatus according to claim 1 or 2, wherein the updated machine learning model is further validated or modified before being transmitted to one or more second network nodes.

4. The apparatus according to claim 1, which further transmits a set of constraints for updating the machine learning model.

5. The apparatus according to claim 1, wherein it further transmits information regarding a reference training procedure for updating the machine learning model, the information regarding the reference training procedure includes at least a set of parameters used to construct the reference training procedure.

6. The apparatus according to claim 1, wherein the message includes the updated machine learning model.

7. The apparatus according to claim 1, wherein the message includes an updated set of weights and biases associated with the updated machine learning model.

8. The apparatus according to claim 1, further obtaining the updated machine learning model by updating the machine learning model based at least partially on the second training data.

9. The apparatus according to claim 1, further comprising the apparatus receiving information indicating a set of constraints for updating the machine learning model.

10. The apparatus according to claim 1, further comprising transmitting the updated machine learning model to one or more first network nodes.

11. The apparatus according to claim 1, wherein the one or more first network nodes comprise a plurality of network nodes of different types.

12. The apparatus according to claim 1, wherein the first training data includes at least one of reference signal measurement information measured at one or more first network nodes, emulated reference signal measurement information, or simulated reference signal measurement information associated with one or more first network nodes.

13. The apparatus according to claim 1, wherein the second training data includes reference signal measurement information measured at a third network node different from the one or more first network nodes.

14. The apparatus according to claim 12 or 13, wherein the reference signal measurement information includes at least channel impulse response measurements taken from one or more received positioning reference signals.

15. A device comprising at least one processor and at least one memory for storing instructions, wherein when an instruction is executed by the at least one processor, the device provides at least one Receiving information comprising at least one of a machine learning model for positioning, a request to update the machine learning model in the device based on second training data, or a request to provide the second training data for updating the machine learning model, wherein the machine learning model is trained based on first training data associated with one or more first network nodes, and receiving To acquire the second training data, Sending at least one of the following: a message indicating the updated machine learning model, or the second training data; The device receives information indicating a set of constraints for updating the machine learning model. A device that performs an action.

16. A device comprising at least one processor and at least one memory for storing instructions, wherein when an instruction is executed by the at least one processor, the device provides at least one Receiving information comprising at least one of a machine learning model for positioning, a request to update the machine learning model in the device based on second training data, or a request to provide the second training data for updating the machine learning model, wherein the machine learning model is trained based on first training data associated with one or more first network nodes, and receiving To acquire the second training data, Sending at least one of the following: a message indicating the updated machine learning model, or the second training data; The device receives information regarding the reference training procedure in order to update the machine learning model. Have them do it, The information relating to the aforementioned reference training procedure includes, in an apparatus, at least a set of parameters used to construct the aforementioned reference training procedure.

17. A device comprising at least one processor and at least one memory for storing instructions, wherein when an instruction is executed by the at least one processor, the device provides at least one Receiving information comprising at least one of a machine learning model for positioning, a request to update the machine learning model in the device based on second training data, or a request to provide the second training data for updating the machine learning model, wherein the machine learning model is trained based on first training data associated with one or more first network nodes, and receiving To acquire the second training data, Sending at least one of the following: a message indicating the updated machine learning model, or the second training data. Have them do it, The aforementioned message is from the device, including the updated machine learning model.

18. A device comprising at least one processor and at least one memory for storing instructions, wherein when an instruction is executed by the at least one processor, the device provides at least one Receiving information comprising at least one of a machine learning model for positioning, a request to update the machine learning model in the device based on second training data, or a request to provide the second training data for updating the machine learning model, wherein the machine learning model is trained based on first training data associated with one or more first network nodes, and receiving To acquire the second training data, Sending at least one of the following: a message indicating the updated machine learning model, or the second training data. Have them do it, The message includes an updated set of weights and biases associated with the updated machine learning model.

19. A device comprising at least one processor and at least one memory for storing instructions, wherein when an instruction is executed by the at least one processor, the device provides at least one Receiving information comprising at least one of a machine learning model for positioning, a request to update the machine learning model in the device based on second training data, or a request to provide the second training data for updating the machine learning model, wherein the machine learning model is trained based on first training data associated with one or more first network nodes, and receiving To acquire the second training data, Sending at least one of the following: a message indicating the updated machine learning model, or the second training data; To send a set of constraints for updating the aforementioned machine learning model. A device that performs an action.

20. The apparatus according to any one of claims 15 to 19, further obtaining the updated machine learning model by updating the machine learning model based on the second training data.

21. The apparatus according to any one of claims 15 to 19, wherein the message is transmitted based on the estimated performance improvement of the updated machine learning model exceeding a threshold.

22. The apparatus according to claim 20, wherein the message is transmitted based on the estimated performance improvement of the updated machine learning model exceeding a threshold.

23. The apparatus according to any one of claims 15 to 19, wherein the first training data includes at least one of reference signal measurement information measured at one or more first network nodes, emulated reference signal measurement information, or simulated reference signal measurement information associated with one or more first network nodes.

24. The apparatus according to any one of claims 15 to 19, wherein the second training data includes reference signal measurement information measured by the apparatus.

25. The apparatus according to claim 23, wherein the reference signal measurement information includes at least channel impulse response measurements taken from one or more received positioning reference signals.

26. The apparatus according to claim 24, wherein the reference signal measurement information includes at least channel impulse response measurements measured from one or more received positioning reference signals.

27. A method for communication, The acquisition of a machine learning model for positioning by a device, wherein the machine learning model is trained on first training data associated with one or more first network nodes, and the acquisition of the machine learning model. The device transmits information including at least one of the following: the machine learning model, a request to update the machine learning model based on second training data, or a request to provide the device with the second training data for updating the machine learning model. The device receives at least one of the following: a message indicating an updated machine learning model, or the second training data. The device transmits the updated machine learning model to at least one or more second network nodes. Methods that include...

28. A non-temporary computer-readable medium containing program instructions, wherein, when executed by a device, the device receives at least: The acquisition of a machine learning model for positioning, wherein the machine learning model is trained on first training data associated with one or more first network nodes. Transmitting information including at least one of the following: the machine learning model, a request to update the machine learning model based on second training data, or a request to provide the second training data for updating the machine learning model in the device; Receiving at least one of a message indicating an updated machine learning model, or the second training data, Transmit the updated machine learning model to at least one or more second network nodes. A non-temporary computer-readable medium that enables the execution of [the action].

Images