Predicting wireless measurements based on virtual access points
Through virtual access point and machine learning models, multi-path component data is used to optimize wireless channel prediction, which solves the problem of degradation of UE positioning accuracy in multi-path environments, and achieves more accurate wireless communication positioning.
Patent Information
- Application Number
- CN202380029788.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2022-04-04
- Filing Date
- 2023-02-28
- Publication Date
- 2024-11-08
AI Technical Summary
In a multipath environment, it is difficult for the prior art to accurately predict the location of user equipment (UE) in wireless communications, especially when the reflective surface is present, traditional positioning methods rely on LoS measurements and lead to a decrease in positioning accuracy.
Virtual access point (VAP) is used to predict wireless channel data, and the machine learning model uses the optimization of multipath component data and real value position to generate predicted wireless measurement output, including parameters such as time of flight (ToF) and signal power, to Determine the location of the UE.
The UE positioning accuracy in a multi-path environment is improved, and the positioning capability of the wireless communication network is enhanced. Especially when the LoS path is blocked, it can effectively use single measurements such as AoA and ToF for positioning.
Smart Images

Figure CN118922736A_ABST
Abstract
Description
[0001] Related Applications
[0002] This application claims the benefit of U.S. Provisional Application No. 17 / 657,920, filed on April 4, 2022, entitled “PREDICTING WIRELESS MEASUREMENTS BASED ON VIRTUAL ACCESS POINTS,” which is assigned to the assignee of this application and is incorporated herein by reference in its entirety. Background Art 1. Technical Field
[0003] The present disclosure relates generally to the field of wireless communications, and more particularly to using virtual access points to predict wireless measurements associated with user equipment (UE) to improve positioning of the UE.
[0004] 2. Related technologies
[0005] Learning the location of a virtual transmitter (e.g., a virtual access point) is one way to learn the wireless environment, including the location of a UE in a multipath environment. A multipath environment may include one or more reflective surfaces, such as building walls, the ground, or other objects that prevent line-of-sight (LoS) transmission of signals between a transmitter and a receiver (e.g., a UE).
[0006] In addition, machine learning has enhanced the accuracy and utility of certain applications. By using training data to train a machine learning model, the model can predict expected information given new inputs. The training data may include "ground truth," which is information known to be true or real and is obtained through observation and measurement, but is not necessarily true. Thus, data containing, for example, location and multipath components may be used to train a machine learning model to generate inferences about the multipath components, which may indicate the location of the UE. Summary of the invention
[0007] In a first aspect of the present disclosure, a method for predicting wireless channel data associated with a user equipment (UE) of a wireless network is disclosed. In some embodiments, the method includes: obtaining a position of the UE; and given the position of the UE, using a first machine learning model to generate an output including one or more predicted wireless measurements, the first machine learning model being trained by at least the following operations: obtaining a training data set including (i) multipath component data and (ii) a true value position of a wireless device, the multipath component data corresponding to a virtual access point associated with an access point of the wireless network, the virtual access point including a position determined based on the position of the access point; and performing optimization with respect to the multipath component data and the true value position.
[0008] In another aspect of the present disclosure, a network node within a wireless network is disclosed. In some embodiments, the network node includes: at least one wireless communication interface; a memory; and one or more processors, the one or more processors are communicatively coupled to the at least one wireless communication interface and the memory, and the one or more processors are configured to: obtain a position of a user equipment (UE); and, given the position of the UE, use a first machine learning model to generate an output including one or more predicted wireless measurements; wherein the first machine learning model is trained by at least the following operations: obtain a training data set including (i) multipath component data and (ii) a true value position of a wireless device, the multipath component data corresponding to a virtual access point associated with an access point of the wireless network, the virtual access point including a position determined based on the position of the access point; and perform optimization with respect to the multipath component data and the true value position.
[0009] In another aspect of the present disclosure, a computer-readable device is disclosed. In some embodiments, the computer-readable device includes a storage medium including a plurality of instructions configured to cause a computerized device to: obtain a position of a user equipment (UE) within a wireless network; and, given the position of the UE, use a first machine learning model to generate an output including one or more predicted wireless measurements; wherein the first machine learning model is trained by at least the following operations: obtaining a training data set including (i) multipath component data and (ii) a true value position of a wireless device, the multipath component data corresponding to a virtual access point associated with an access point of the wireless network, the virtual access point including a position determined based on the position of the access point; and performing optimization with respect to the multipath component data and the true value position.
[0010] In another aspect of the present disclosure, a computerized device is disclosed. In some embodiments, the computerized device includes: a component for obtaining a location of a user equipment (UE) of a wireless network; and a component for generating an output including one or more predicted wireless measurements using a first machine learning model given the location of the UE, the first machine learning model being trained by utilizing at least the following items: a component for obtaining a training data set including (i) multipath component data and (ii) a true value location of a wireless device, the multipath component data corresponding to a virtual access point associated with an access point of the wireless network, the virtual access point including a location determined based on the location of the access point; and a component for performing optimization with respect to the multipath component data and the true value location.
[0011] This summary is neither intended to identify key features or essential features of the claimed subject matter nor is it intended to be used alone to determine the scope of the claimed subject matter. The subject matter should be understood by reference to appropriate portions of the entire specification of the disclosure, any or all of the drawings, and each claim. The foregoing and other features and examples will be described in more detail in the following description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0012] Figure 1 is a diagram of a positioning system according to an embodiment.
[0013] Figure 2 is a diagram of a fifth generation (5G) new radio (NR) positioning system, which illustrates a positioning system implemented within a 5G NR communication system (e.g., Figure 1 implementation scheme of a positioning system).
[0014] Figure 3A is a diagram illustrating the relative locations of a physical transmitter, a virtual transmitter, and a user equipment (UE) according to some embodiments.
[0015] Figure 3B is a diagram illustrating the relative positions of a physical transmitter, a virtual transmitter, and multiple UEs according to some embodiments.
[0016] Figure 3C is a diagram illustrating the relative positions of a physical transmitter, multiple virtual transmitters, and a UE according to some embodiments.
[0017] Figure 4 is a block diagram illustrating the collection of wireless channel measurement data according to some embodiments.
[0018] Figure 5 is a block diagram illustrating a mechanism for training a machine learning model according to some embodiments.
[0019] Fig. 6A is a block diagram illustrating inference of wireless channel measurements based on a trained machine learning model according to some embodiments.
[0020] Figure 6B is another block diagram illustrating inference of wireless channel measurements based on a trained machine learning model according to some embodiments.
[0021] Figure 6C is a block diagram illustrating the use of classifiers and regressors in a machine learning model to predict and obtain wireless measurements according to some embodiments.
[0022] Figure 7 is a flow chart of a method of predicting wireless channel data associated with a UE of a wireless network according to some embodiments.
[0023] Figure 8 is a flowchart of a method for training a machine learning model according to some embodiments.
[0024] Fig. 9 is a flow chart of another method of predicting wireless channel data associated with a UE of a wireless network according to some embodiments.
[0025] Fig.10 is a block diagram of an implementation of a UE that may be utilized in implementations as described herein.
[0026] Fig.11 is a block diagram of an implementation of a base station that may be utilized in implementations as described herein.
[0027] Fig.12 is a block diagram of an embodiment of a computer system that may be utilized in embodiments as described herein.
[0028] Similar reference symbols in the various drawings indicate similar elements according to certain example implementations. In addition, multiple instances of an element can be indicated by adding a letter or hyphen and a second number after the first number of the element. For example, multiple instances of element 110 can be indicated as 110-1, 110-2, 110-3, etc. or 110a, 110b, 110c, etc. When only the first number is used to refer to such an element, any instance of the element should be understood (for example, the element 110 in the previous example will refer to elements 110-1, 110-2 and 110-3 or to elements 110a, 110b and 110c). DETAILED DESCRIPTION
[0029] The following description is directed to certain specific implementations for the purpose of describing the innovative aspects of each embodiment. However, one of ordinary skill in the art will readily recognize that the teachings herein can be applied in a variety of different ways. The described specific implementations can be implemented in any device, system, or network capable of sending and receiving radio frequency (RF) signals according to any communication standard, such as any of the Institute of Electrical and Electronics Engineers (IEEE) IEEE 802.11 standards (including those identified as Technical standards), Standard, Code Division Multiple Access (CDMA), Frequency Division Multiple Access (FDMA), Time Division Multiple Access (TDMA), Global System for Mobile Communications (GSM), GSM / General Packet Radio Service (GPRS), Enhanced Data GSM Environment (EDGE), Terrestrial Trunked Radio (TETRA), Wideband CDMA (W-CDMA), Evolution-Data Optimized (EV-DO), 1xEV-DO, EV-DO Revision A, EV-DO Revision B, High Rate Packet Data (HRPD), High Speed Packet Access (HSPA), High Speed Downlink Packet Access (HSDPA), High Speed Uplink Packet Access (HSUPA), Evolved High Speed Packet Access (HSPA+), Long Term Evolution (LTE), Advanced Mobile Phone System (AMPS), or other known signals for communicating within a wireless, cellular, or Internet of Things (IoT) network, such as a system utilizing 3G, 4G, 5G, 6G or further implementations thereof.
[0030] As used herein, "module" may refer to at least a portion of a computer executable instruction. In some embodiments, a module may be implemented by a hardware processor configured to execute the corresponding computer executable instruction. The hardware processor may include an integrated circuit device associated with a computing device (such as a server or user equipment (e.g., UE, desktop computer, laptop computer, tablet computer, mobile phone, etc.)), which may be programmed to perform specific tasks. In some embodiments, multiple modules may be implemented as a single module. In some embodiments, a single module may be implemented as multiple modules. In some embodiments, two or more modules may be executed by the same device (e.g., the same server, the same computing device).
[0031] As used herein, "RF signals" include electromagnetic waves that transmit information through the space between a transmitter (or transmitting device) and a receiver (or receiving device). As used herein, a transmitter may transmit a single "RF signal" or multiple "RF signals" to a receiver. However, due to the propagation characteristics of each RF signal through multiple channels or paths, a receiver may receive multiple "RF signals" corresponding to each transmitted RF signal.
[0032] Additionally, unless otherwise noted, references to "reference signals," "positioning reference signals," "reference signals for positioning," etc. may be used to refer to signals used to locate user equipment (UE). As described in more detail herein, such signals may include any of a variety of signal types, but may not necessarily be limited to positioning reference signals (PRS) as defined in relevant wireless standards.
[0033] Figure 1 is a simplified illustration of a positioning system 100 according to an embodiment, in which a UE 105, a location server 160, and / or other components of the positioning system 100 may use the techniques provided herein for, for example, predicting wireless measurements of the UE 105. The techniques described herein may be implemented by one or more components of the positioning system 100. The positioning system 100 may include: a UE 105; one or more satellites 110 (also referred to as spacecraft (SVs)) for a global navigation satellite system (GNSS) such as a global positioning system (GPS), GLONASS, Galileo, or BeiDou; a base station 120; an access point (AP) 130; a location server 160; a network 170; and an external client 180. In general, the positioning system 100 may estimate the location of the UE 105 based on RF signals received by and / or sent from the UE 105 and the known locations of other components (e.g., GNSS satellites 110, base stations 120, APs 130) that send and / or receive RF signals. Reference Figure 2 Additional details regarding specific location estimation techniques are discussed in more detail.
[0034] It should be noted that Figure 1 Only a generalized illustration of various components is provided, any or all of which may be utilized as appropriate, and each component may be repeated as needed. Specifically, although only one UE 105 is illustrated, it should be understood that many UEs (e.g., hundreds, thousands, millions, etc.) may utilize the positioning system 100. Similarly, the positioning system 100 may include more than one UE 105. Figure 1 100. The illustrated connections connecting the various components in the positioning system 100 include data and signaling connections, which may include additional (intermediate) components, direct or indirect physical and / or wireless connections, and / or additional networks. In addition, the components may be rearranged, combined, separated, replaced, and / or omitted depending on the desired functionality. In some embodiments, for example, the external client 180 may be directly connected to the location server 160. Those of ordinary skill in the art will recognize many modifications to the illustrated components.
[0035] Depending on the desired functionality, network 170 may include any one of a variety of wireless and / or wired networks. Network 170 may, for example, include any combination of public and / or private networks, local area networks and / or wide area networks, etc. In addition, network 170 may utilize one or more wired and / or wireless communication technologies. In some embodiments, network 170 may include, for example, cellular or other mobile networks, wireless local area networks (WLANs), wireless wide area networks (WWANs), and / or the Internet. Examples of network 170 include long-term evolution (LTE) wireless networks, fifth-generation (5G) wireless networks (also referred to as new radio (NR) wireless networks or 5GNR wireless networks), Wi-Fi WLANs, and the Internet. LTE, 5G, and NR are wireless technologies defined or being defined by the Third Generation Partnership Project (3GPP). Network 170 may also include more than one network and / or more than one type of network.
[0036] The base station 120 and the access point (AP) 130 are communicatively coupled to the network 170. In some embodiments, the base station 120 may be owned, maintained and / or operated by a cellular network provider and may employ any of a variety of wireless technologies, as described below. Depending on the technology of the network 170, the base station 120 may include a node B, an evolved node B (eNodeB or eNB), a transceiver base station (BTS), a radio base station (RBS), an NR node B (gNB), a next generation eNB (ng-eNB), etc. In the case where the network 170 is a 5G network, the base station 120 as a gNB or ng-eNB may be part of a next generation radio access network (NG-RAN) that may be connected to a 5G core network (5GC). In view of the open radio access network (O-RAN) and / or virtualized radio access network (V-RAN or vRAN) in 5G or higher networks, the functionality performed by the base station 120 in the earlier networks (e.g., 3G and 4G) may be divided into different functional components (e.g., radio unit (RU), distributed unit (DU), central unit (CU)) and layers (e.g., L1 / L2 / L3), which may be executed on different devices at different locations connected, for example, via fronthaul connections, midhaul connections, and backhaul connections. As described herein, a "base station" (or ng-eNB, gNB, etc.) may include any or all of these functional components. For example, AP 130 may include a Wi-Fi AP or AP or an AP with cellular capabilities (e.g., 4G LTE and / or 5G NR). Thus, UE 105 can send and receive information with network connectivity devices (such as location server 160) by accessing network 170 via base station 120 using first communication link 133. Additionally or alternatively, because AP 130 can also be communicatively coupled with network 170, UE 105 can communicate with network connectivity and Internet connectivity devices (including location server 160) using second communication link 135 or via one or more other UEs 145.
[0037] As used herein, the term "base station" may generally refer to a single physical transmission point or multiple co-located physical transmission points that may be located at a base station 120. A transmit receive point (TRP) (also referred to as a transmit / receive point) corresponds to this type of transmission point, and the term "TRP" may be used interchangeably with the terms "gNB," "ng-eNB," and "base station" herein. In some cases, a base station 120 may include multiple TRPs—for example, where each TRP is associated with a different antenna or different antenna array of the base station 120. As used herein, the transmit function of a TRP may be performed using a transmit point (TP) and / or the receive function of a TRP may be performed by a receive point (RP), which may be physically separated or different from the TP. That is, a TRP may include both a TP and an RP. A physical transmission point may include an antenna array of a base station 120 (for example, as in a multiple-input multiple-output (MIMO) system and / or where the base station employs beamforming). The term "base station" may additionally refer to multiple non-co-located physical transmission points, which may be a distributed antenna system (DAS) (a network of spatially separated antennas connected to a common source via a transmission medium) or a remote radio head (RRH) (a remote base station connected to a serving base station).
[0038] As used herein, the term "cell" may generally refer to a logical communication entity for communicating with base station 120, and may be associated with an identifier (e.g., a physical cell identifier (PCID), a virtual cell identifier (VCID)) for distinguishing adjacent cells operating via the same or different carriers. In some examples, a carrier may support multiple cells, and different cells may be configured according to different protocol types that may provide access for different types of devices (e.g., machine type communication (MTC), narrowband Internet of Things (NB-IoT), enhanced mobile broadband (eMBB), or other protocols). In some cases, the term "cell" may refer to a portion of a geographic coverage area (e.g., a sector) over which the logical entity operates.
[0039] The location server 160 may include a server and / or other computing device configured to determine an estimated location of the UE 105 and / or provide data (e.g., “assistance data”) to the UE 105 to facilitate location measurement and / or location determination by the UE 105. According to some embodiments, the location server 160 may include a Home Secure User Plane Location (SUPL) Location Platform (H-SLP), which may support a SUPL User Plane (UP) positioning solution defined by the Open Mobile Alliance (OMA), and may support location services for the UE 105 based on subscription information about the UE 105 stored in the location server 160. In some embodiments, the location server 160 may include a Discovery SLP (D-SLP) or an Emergency SLP (E-SLP). The location server 160 may also include an Enhanced Serving Mobile Location Center (E-SMLC), which supports positioning of the UE 105 using a Control Plane (CP) positioning solution for LTE radio access of the UE 105. The location server 160 may further include a location management function (LMF) that supports positioning of the UE 105 using a control plane (CP) positioning solution for NR or LTE radio access of the UE 105 .
[0040] In the CP positioning solution, from the perspective of the network 170, the signaling for controlling and managing the positioning of the UE 105 can use existing network interfaces and protocols and be exchanged as signaling between the elements of the network 170 and with the UE 105. In the UP positioning solution, from the perspective of the network 170, the signaling for controlling and managing the positioning of the UE 105 can be exchanged between the location server 160 and the UE 105 as data (e.g., data transmitted using the Internet Protocol (IP) and / or the Transmission Control Protocol (TCP)).
[0041] As previously mentioned (and discussed in more detail below), the estimated position of UE 105 may be based on measurements of RF signals transmitted from and / or received by UE 105. In particular, these measurements may provide information about the relative distances and / or angles of UE 105 from one or more components in positioning system 100 (e.g., GNSS satellites 110, APs 130, base stations 120). The estimated position of UE 105 may be estimated geometrically (e.g., using multi-angle measurements and / or multilateration) based on the distance and / or angle measurements along with the known positions of the one or more components.
[0042] Although ground components (such as AP 130 and base station 120) can be fixed, the embodiments are not limited thereto. Mobile components can be used. For example, in some embodiments, the position of UE 105 can be estimated based at least in part on the measurement of RF signals 140 communicated between UE 105 and one or more other UEs 145 (the one or more other UEs 145 can be mobile or fixed). When one or more other UEs 145 are used in the positioning determination of a specific UE 105, the UE 105 whose position is to be determined can be referred to as a "target UE", and each UE in the one or more other UEs 145 can be referred to as an "anchor UE". In order to determine the positioning of the target UE, the corresponding positions of the one or more anchor UEs can be known and / or determined jointly with the target UE. Direct communication between the one or more other UEs 145 and UE 105 may include side links and / or similar device-to-device (D2D) communication technologies. The side link defined by 3GPP is a form of D2D communication under the cellular LTE and NR standards.
[0043] The estimated position of UE 105 may be used in various applications, such as to assist a user of UE 105 with direction finding or navigation or to assist another user (e.g., associated with external client 180) in locating UE 105. "Position" is also referred to herein as "position estimate," "estimated position," "position," "position estimate," "position fix," "estimated position," "position fix," or "fix." The process of determining a position may be referred to as "positioning," "position determination," "position determination," or the like. The position of UE 105 may include an absolute position of UE 105 (e.g., latitude and longitude, and possibly altitude) or a relative position of UE 105 (e.g., a position expressed as a distance north or south, east or west, and possibly above or below some other known fixed position (including, for example, the position of base station 120 or AP 130) or some other position (such as the position of UE 105 at some known previous time, or the position of another UE 145 at some known previous time)). The location may be specified as a geodetic location including coordinates that may be absolute (e.g., latitude, longitude, and optionally altitude), relative (e.g., relative to some known absolute location), or local (e.g., X, Y, and optionally Z coordinates according to a coordinate system defined relative to a local area, such as a factory, warehouse, university campus, shopping mall, stadium, or convention center). The location may alternatively be a city location, and may then include one or more of a street address (e.g., including a country, state, county, city, road and / or street name or label, and / or road or street number) and / or a label or name of a place, a building, a portion of a building, a floor of a building, and / or a room within a building, etc. The location may also include an uncertainty or error indication, such as a horizontal and possibly vertical distance that the location is expected to have error, or an indication of an area or volume (e.g., a circle or ellipse) within which the UE 105 is expected to be located with a certain confidence level (e.g., 95% confidence).
[0044] The external client 180 may be a web server or remote application that may have some association with the UE 105 (e.g., may be accessed by a user of the UE 105), or may be a server, application, or computer system that provides location services to one or more other users, which may include obtaining and providing the location of the UE 105 (e.g., to enable services such as friend or relative locating, or child or pet locating). Additionally or alternatively, the external client 180 may obtain the location of the UE 105 and provide it to emergency service providers, government agencies, etc.
[0045] As previously mentioned, the example positioning system 100 may be implemented using a wireless communication network, such as an LTE-based or 5G NR-based network. Figure 2A diagram of a 5G NR positioning system 200 is shown, which illustrates an embodiment of a positioning system (e.g., positioning system 100) that implements 5G NR. The 5G NR positioning system 200 can be configured to determine the location of the UE 105 by using access nodes to implement one or more positioning methods, and the access nodes can include NR B nodes (gNBs) 210-1 and 210-2 (collectively referred to as gNBs 210 in this document), ng-eNBs 214, and / or WLANs 216. The gNBs 210 and / or ng-eNBs 214 can communicate with the UE 105 to determine the location of the UE 105. Figure 1 The base station 120 corresponds to the WLAN 216, and the WLAN 216 can be connected to the Figure 1 130. Optionally, the 5G NR positioning system 200 may also be configured to determine the location of the UE 105 by using the LMF 220 (which may correspond to the location server 160) to implement one or more positioning methods. Here, the 5G NR positioning system 200 includes the UE 105, and components of the 5G NR network, including the next generation (NG) radio access network (RAN) (NG-RAN) 235 and the 5G core network (5GCN) 240. The 5G network may also be referred to as an NR network; the NG-RAN 235 may be referred to as a 5G RAN or an NR RAN; and the 5G CN 240 may be referred to as an NG core network. The 5G NR positioning system 200 may further utilize information from GNSS satellites 110 of a GNSS system such as the Global Positioning System (GPS) or a similar system such as GLONASS, Galileo, BeiDou, Indian Regional Navigation Satellite System (IRNSS). Additional components of the 5G NR positioning system 200 are described below. The 5G NR positioning system 200 may include additional or alternative components.
[0046] It should be noted that Figure 2 Only generalized illustrations of various components are provided, any or all of which may be utilized as appropriate, and each component may be repeated or omitted as desired. Specifically, although only one UE 105 is illustrated, it will be understood that many UEs (e.g., hundreds, thousands, millions, etc.) may utilize the 5G NR positioning system 200. Similarly, the 5G NR positioning system 200 may include a larger (or smaller) number of GNSS satellites 110, gNB 210, ng-eNB 214, wireless local area network (WLAN) 216, access and mobility management function (AMF) 215, external client 230, and / or other components. The illustrated connections connecting the various components in the 5G NR positioning system 200 include data and signaling connections, which may include additional (intermediate) components, direct or indirect physical and / or wireless connections, and / or additional networks. In addition, the components may be rearranged, combined, separated, replaced, and / or omitted depending on the desired functionality.
[0047] UE 105 may include and / or be referred to as a device, a mobile device, a wireless device, a mobile terminal, a terminal, a mobile station (MS), a terminal (SET) enabled with secure user plane location (SUPL), or some other name. In addition, UE 105 may correspond to a cellular phone, a smart phone, a laptop computer, a tablet device, a personal data assistant (PDA), a navigation device, an Internet of Things (IoT) device, or some other portable or mobile device. Typically, although not necessarily, UE 105 may support the use of one or more radio access technologies (RATs) such as GSM, CDMA, W-CDMA, LTE, High Speed Packet Data (HRPD), IEEE 802.11 Bluetooth, Worldwide Interoperability for Microwave Access (WiMAX TM ), 5G NR (e.g., using NG-RAN 235 and 5GCN 240). UE 105 may also support wireless communications using WLAN 216, which is similar to one or more RATs and as previously described with respect to Figure 1 The use of one or more of these RATs may allow UE 105 (e.g., via Figure 2 240, or possibly via a Gateway Mobile Location Center (GMLC) 225) to communicate with an external client 230 and / or allow an external client 230 (e.g., via GMLC 225) to receive location information about UE 105. When implemented in or communicatively coupled with a 5G NR network, Figure 2 The external client 230 may correspond to Figure 1 External client 180.
[0048] UE 105 may include a single entity or may include multiple entities, such as in a personal area network in which a user may employ audio, video, and / or data I / O devices, and / or body sensors and separate wired or wireless modems. The estimate of the location of UE 105 may be referred to as location, location estimate, location fix, fix, position, position estimate, or position fix, and may be geodetic, providing location coordinates (e.g., latitude and longitude) about UE 105, which may or may not include an altitude component (e.g., altitude; height above or depth below ground level, floor level, or basement level). Alternatively, the location of UE 105 may be expressed as a municipal location (e.g., expressed as a postal address or a designation of a point or small area in a building (such as a specific room or floor)). The location of UE 105 may also be expressed as an area or volume (geographically or municipally defined) within which UE 105 is expected to be located with a certain probability or confidence level (e.g., 67%, 95%, etc.). The location of the UE 105 may also be a relative location, including, for example, a distance and direction or relative X, Y (and Z) coordinates defined relative to some origin at a known location, which may be defined geodetically, municipally, or with reference to a point, area, or volume indicated on a map, floor plan, or building plan. In the descriptions contained herein, the use of the term "location" may include any of these variations unless otherwise indicated. When calculating the location of the UE, local X, Y, and possibly Z coordinates are typically solved, and then the local coordinates are converted to absolute coordinates (e.g., with respect to latitude, longitude, and altitude above or below mean sea level) if necessary.
[0049] Figure 2 The base stations in the NG-RAN 235 shown in FIG. 2 may correspond to Figure 1 The base stations 120 in the NG-RAN 235 may include gNBs 210. Pairs of gNBs 210 in the NG-RAN 235 may be connected to each other (e.g., Figure 2 The communication interface between the base stations (gNB 210 and / or ng-eNB 214) may be referred to as an Xn interface 237. Access to the 5G network is provided to the UE 105 via wireless communication between the UE 105 and one or more gNBs 210, which may provide wireless communication access to the 5G CN 240 on behalf of the UE 105 using 5G NR. The wireless interface between the base station (gNB 210 and / or ng-eNB 214) and the UE 105 may be referred to as an Uu interface 239. 5G NR radio access may also be referred to as NR radio access or 5G radio access. Figure 2, it is assumed that the serving gNB for UE 105 is gNB 210-1, but other gNBs (e.g., gNB 210-2) may act as serving gNBs if UE 105 moves to another location, or may act as secondary gNBs to provide additional throughput and bandwidth to UE 105.
[0050] Figure 2 The base stations in the NG-RAN 235 shown in the figure may additionally or alternatively include a next generation evolved Node B (also referred to as ng-eNB) 214. The ng-eNB 214 may be connected to one or more gNBs 210 in the NG-RAN 235—e.g., directly or indirectly via other gNBs 210 and / or other ng-eNBs. The ng-eNB 214 may provide LTE radio access and / or evolved LTE (eLTE) radio access to the UE 105. Figure 2 Some of the gNBs 210 (e.g., gNB 210-2) and / or ng-eNBs 214 in the 5G CN 240 may be configured to function as positioning-only beacons that may transmit signals (e.g., positioning reference signals (PRS)) and / or may broadcast assistance data to assist in positioning the UE 105, but may not receive signals from the UE 105 or from other UEs. Some of the gNBs 210 (e.g., gNB 210-2 and / or another gNB not shown) and / or ng-eNBs 214 may be configured to function as detection-only nodes that may scan for signals containing, for example, PRS data, assistance data, or other location data. Such detection-only nodes may not transmit signals or data to the UE, but may transmit signals or data (relating to, for example, PRS, assistance data, or other location data) to other network entities (e.g., one or more components of the 5G CN 240, an external client 230, or a controller), which may receive and store the data or use the data to locate at least the UE 105. Note that although in Figure 2 Only one ng-eNB 214 is shown in the figure, but some embodiments may include multiple ng-eNBs 214. Base stations (e.g., gNBs 210 and / or ng-eNB 214) may communicate directly with each other via an Xn communication interface. Additionally or alternatively, the base station may communicate directly or indirectly with other components of the 5G NR positioning system 200 (such as LMF 220 and AMF 215).
[0051] The 5G NR positioning system 200 may also include one or more WLANs 216 that may be connected to a non-3GPP interworking function (N3IWF) 250 in the 5GCN 240 (e.g., in the case of an untrusted WLAN 216). For example, the WLAN 216 may support IEEE 802.11 Wi-Fi access for the UE 105 and may include one or more Wi-Fi APs (e.g., Figure 1 130). Here, the N3IWF 250 may be connected to other elements in the 5G CN 240, such as the AMF 215. In some embodiments, the WLAN 216 may support another RAT, such as Bluetooth. The N3IWF 250 may provide support for secure access of the UE 105 to other elements in the 5G CN 240 and / or may support interworking of one or more protocols used by the WLAN 216 and the UE 105 with one or more protocols used by other elements of the 5GCN 240 (such as the AMF 215). For example, the N3IWF 250 may support: establishing an IPSec tunnel with the UE 105, terminating the IKEv2 / IPSec protocol with the UE 105, terminating the N2 and N3 interfaces to the 5G CN 240 for the control plane and the user plane, respectively, and relaying uplink (UL) and downlink (DL) control plane non-access stratum (NAS) signaling across the N1 interface between the UE 105 and the AMF 215. In some other embodiments, the WLAN 216 may be directly connected to elements in the 5G CN 240 (e.g., Figure 2 215) and does not pass through the N3IWF 250. For example, the direct connection of the WLAN 216 to the 5GCN 240 may occur when the WLAN 216 is a trusted WLAN for the 5GCN 240 and may use a trusted WLAN interworking function (TWIF) ( Figure 2 (not shown). Note that although Figure 2 Only one WLAN 216 is shown in FIG. 2 , but some embodiments may include multiple WLANs 216 .
[0052] The access node may include any of a variety of network entities that enable communication between the UE 105 and the AMF 215. As mentioned, this may include a gNB 210, an ng-eNB 214, a WLAN 216, and / or other types of cellular base stations. However, the access node providing the functionality described herein may additionally or alternatively include a cellular network entity that enables communication with the UE 105. Figure 2An entity that communicates with any of a variety of RATs not illustrated in the specification (which may include non-cellular technologies). Therefore, as used in the embodiments described below, the term "access node" may include but is not necessarily limited to gNB 210, ng-eNB 214, or WLAN 216.
[0053] In some embodiments, an access node (such as gNB 210, ng-eNB 214, and / or WLAN 216) (alone or in combination with other components of 5G NR positioning system 200) may be configured to: in response to receiving a request for location information from LMF 220, obtain a location measurement for an uplink (UL) signal received from UE 105 and / or obtain a DL location measurement from UE 105 for a downlink (DL) signal received by UE 105 from one or more access nodes. As mentioned, although Figure 2 The access nodes (gNB 210, ng-eNB 214, and WLAN 216) are depicted as being configured to communicate in accordance with 5G NR, LTE, and Wi-Fi communication protocols, respectively, but access nodes configured to communicate in accordance with other communication protocols may be used, such as, for example, a Node B using a Wideband Code Division Multiple Access (WCDMA) protocol for a Universal Mobile Telecommunications Service (UMTS) Terrestrial Radio Access Network (UTRAN), an eNB using an LTE protocol for an Evolved UTRAN (E-UTRAN), or an eNB using a LTE protocol for a WLAN. For example, in a 4G Evolved Packet System (EPS) that provides LTE wireless access to UE 105, the RAN may include an E-UTRAN, which may include base stations including eNBs that support LTE wireless access. The core network for the EPS may include an Evolved Packet Core (EPC). The EPS may then include the E-UTRAN plus the EPC, where Figure 2 , E-UTRAN corresponds to NG-RAN 235 and EPC corresponds to 5GCN 240. The methods and techniques described herein for obtaining the municipal location of UE 105 may be applicable to such other networks.
[0054] The gNB 210 and ng-eNB 214 may communicate with the AMF 215, which communicates with the LMF 220 for positioning functionality. The AMF 215 may support mobility of the UE 105, including cell changes and handovers of the UE 105 from an access node of a first RAT (e.g., a gNB 210, ng-eNB 214, or WLAN 216) to an access node of a second RAT. The AMF 215 may also be involved in supporting signaling connections to the UE 105 and possibly supporting data and voice bearers for the UE 105. The LMF 220 may support positioning the UE 105 using a CP positioning solution when the UE 105 accesses the NG-RAN 235 or the WLAN 216, and may support various positioning procedures and methods, including UE-assisted / UE-based and / or network-based procedures / methods, such as Assisted GNSS (A-GNSS), Observed Time Difference of Arrival (OTDOA) (which may be referred to as Time Difference of Arrival (TDOA) in NR), Real-Time Kinematics (RTK), Precise Point Positioning (PPP), Differential GNSS (DGNSS), Enhanced Cell ID (ECID), Angle of Arrival (AoA), Angle of Departure (AoD), WLAN positioning, Round Trip Signal Propagation Delay (RTT), Multi-Cell RTT, and / or other positioning procedures and methods. The LMF 220 may also process a location service request for the UE 105 received, for example, from the AMF 215 or from the GMLC 225. The LMF 220 may be connected to the AMF 215 and / or the GMLC 225. In some embodiments, the network (such as 5GCN 240) may additionally or alternatively implement other types of location support modules, such as an evolved serving mobile location center (E-SMLC) or a SUPL location platform (SLP). Note that in some embodiments, at least a portion of the positioning functionality (including determining the location of UE 105) can be performed at UE 105 (e.g., by measuring downlink PRS (DL-PRS) signals sent by wireless nodes (such as gNB 210, ng-eNB 214 and / or WLAN 216) and / or using assistance data provided to UE 105 by, for example, LMF 220).
[0055] The Gateway Mobile Location Center (GMLC) 225 may support location requests for the UE 105 received from the external client 230, and may forward such location requests to the AMF 215 for forwarding by the AMF 215 to the LMF 220. A location response (e.g., including a location estimate for the UE 105) from the LMF 220 may similarly be returned to the GMLC 225 directly or via the AMF 215, and the GMLC 225 may then return the location response (e.g., including the location estimate) to the external client 230.
[0056] A network exposure function (NEF) 245 may be included in the 5GCN 240. The NEF 245 may support secure exposure of capabilities and events regarding the 5GCN 240 and the UE 105 to the external clients 230, which may therefore be referred to as access functions (AFs) and may enable secure provisioning of information from the external clients 230 to the 5GCN 240. The NEF 245 may be connected to the AMF 215 and / or the GMLC 225 for the purpose of obtaining the location (e.g., municipal location) of the UE 105 and providing the location to the external clients 230.
[0057] like Figure 2 As further illustrated in , LMF 220 may communicate with gNB 210 and / or with ng-eNB 214 using NR Positioning Protocol Annex (NRPPa) as defined in 3GPP Technical Specification (TS) 38.455. NRPPa messages may be communicated between gNB 210 and LMF 220 and / or between ng-eNB 214 and LMF 220 via AMF 215. Figure 2 As further illustrated in , the LMF 220 and the UE 105 may communicate using the LTE Positioning Protocol (LPP) as defined in 3GPP TS 37.355. Here, LPP messages may be passed between the UE 105 and the LMF 220 via the AMF 215 and the serving gNB 210-1 or serving ng-eNB 214 of the UE 105. For example, the LPP messages may be passed between the LMF 220 and the AMF 215 using messages for service-based operations (e.g., based on the Hypertext Transfer Protocol (HTTP)), and may be passed between the AMF 215 and the UE 105 using the 5G NAS protocol. The LPP protocol may be used to support positioning of the UE 105 using UE-assisted and / or UE-based positioning methods such as A-GNSS, RTK, TDOA, multi-cell RTT, AoD, and / or ECID. The NRPPa protocol may be used to support positioning of UE 105 using network-based positioning methods such as ECID, AoA, uplink TDOA (UL-TDOA) and / or may be used by LMF 220 to obtain location-related information from gNB 210 and / or ng-eNB 214, such as parameters defining DL-PRS transmissions from gNB 210 and / or ng-eNB 214.
[0058] In the case where the UE 105 accesses the WLAN 216, the LMF 220 may use NRPPa and / or LPP to obtain the location of the UE 105 in a manner similar to that just described for the UE 105 accessing the gNB 210 or ng-eNB 214. Thus, NRPPa messages may be passed between the WLAN 216 and the LMF 220 via the AMF 215 and the N3IWF 250 to support network-based positioning of the UE 105 and / or to pass other location information from the WLAN 216 to the LMF 220. Alternatively, NRPPa messages may be passed between the N3IWF 250 and the LMF 220 via the AMF 215 to support network-based positioning of the UE 105 based on location-related information and / or location measurements that are known or accessible to the N3IWF 250 and passed from the N3IWF 250 to the LMF 220 using NRPPa. Similarly, LPP and / or LPP messages may be passed between the UE 105 and the LMF 220 via the AMF 215, the N3IWF 250, and the serving WLAN 216 of the UE 105 to support UE-assisted or UE-based positioning of the UE 105 by the LMF 220.
[0059] In the 5G NR positioning system 200, positioning methods can be classified as "UE-assisted" or "UE-based". This may depend on where the request to determine the location of the UE 105 originates. For example, if the request originates from the UE (e.g., from an application or "app" executed by the UE), the positioning method can be classified as UE-based. On the other hand, if the request originates from an external client or AF 230, LMF 220, or other device or service within the 5G network, the positioning method can be classified as UE-assisted (or "network-based").
[0060] With UE-assisted positioning methods, the UE 105 may obtain location measurements and send these measurements to a location server (e.g., LMF 220) for use in computing a location estimate for the UE 105. For RAT-dependent positioning methods, the location measurements may include one or more of the following for one or more access points of the gNB 210, ng-eNB 214, and / or WLAN 216: received signal strength indicator (RSSI), round trip signal propagation time (RTT), reference signal received power (RSRP), reference signal received quality (RSRQ), reference signal time difference (RSTD), time of arrival (TOA), AoA, receive time-transmit time difference (Rx-Tx), differential AoA (DAoA), AoD, or timing advance (TA). Additionally or alternatively, similar measurements may be made on sidelink signals transmitted by other UEs, which may be used as anchor points for positioning the UE 105 if the locations of these other UEs are known. The position measurements may additionally or alternatively include measurements for a RAT-independent positioning method, such as GNSS (eg, GNSS pseudoranges with respect to GNSS satellites 110 , GNSS code phase, and / or GNSS carrier phase), WLAN, and the like.
[0061] Using the UE-based positioning method, the UE 105 can obtain a position measurement (e.g., which can be the same or similar to the position measurement of the UE-assisted positioning method), and the position of the UE 105 can be further calculated (e.g., with the aid of auxiliary data received from a location server such as LMF 220, SLP or broadcast by gNB 210, ng-eNB 214, or WLAN 216).
[0062] Using a network-based positioning method, one or more base stations (e.g., gNB 210 and / or ng-eNB 214), one or more APs (e.g., APs in WLAN 216), or N3IWF 250 may obtain location measurements (e.g., measurements of RSSI, RTT, RSRP, RSRQ, AOA, or TOA) of signals sent by UE 105, and / or may receive measurements obtained by UE 105 or, in the case of N3IWF250, by APs in WLAN 216, and may send these measurements to a location server (e.g., LMF 220) for use in calculating a position estimate for UE 105.
[0063] The positioning of UE 105 may also be classified as UL-based, DL-based, or DL-UL-based according to the type of signal used for positioning. For example, if the positioning is based only on signals received at UE 105 (e.g., from a base station or other UE), the positioning may be classified as DL-based. On the other hand, if the positioning is based only on signals sent by UE 105 (which may be received by, for example, a base station or other UE), the positioning may be classified as UL-based. DL-UL-based positioning includes positioning based on signals sent and received by UE 105, such as RTT-based positioning. Sidelink (SL)-assisted positioning includes signals communicated between UE 105 and one or more other UEs. According to some embodiments, the UL, DL, or DL-UL positioning described herein may be able to use SL signaling as a supplement or replacement for SL, DL, or DL-UL signaling.
[0064] Depending on the type of positioning (e.g., UL-based, DL-based, or DL-UL-based), the type of reference signal used may be different. For example, for DL-based positioning, these signals may include PRS (e.g., DL-PRS sent by a base station or SL-PRS sent by other UEs), which may be used for TDOA, AoD, and RTT measurements. Other reference signals that may be used for positioning (UL, DL, or DL-UL) may include: Sounding Reference Signal (SRS), Channel State Information Reference Signal (CSI-RS), Synchronization Signal (e.g., Synchronization Signal Block (SSB) Synchronization Signal (SS)), Physical Uplink Control Channel (PUCCH), Physical Uplink Shared Channel (PUSCH), Physical Sidelink Shared Channel (PSSCH), Demodulation Reference Signal (DMRS), etc. In addition, the reference signal may be sent in a Tx beam and / or received in an Rx beam (e.g., using beamforming techniques), which may affect angle measurements such as AoD and / or AoA.
[0065] Virtual Access Point
[0066] Figure 3A306. In some embodiments, the physical transmitter 302 may be an example of a base station 120 or an access point 130. In some embodiments, the virtual transmitter may be a "virtual access point" (VAP) 304. As discussed herein, a VAP (also referred to herein as a "virtual anchor") is a virtual non-physical entity that may at least partially describe an RF environment such as a wireless communication network. More specifically, in some cases, a VAP may represent a location where its location is mirrored relative to a reflective surface 308. In some embodiments, the reflective surface 308 may be a surface (such as the floor or wall of a building) that causes electromagnetic (e.g., RF) waves to reflect from it.
[0067] like Figure 3A , VAP 304 is identifiable and positioned as a mirror image of physical transmitter (e.g., gNB) 302 across reflective surface 308. The distance between VAP 304 and reflective surface 308 is h1, and the distance between physical transmitter 302 and reflective surface 308 is h2, where h1 and h2 are equal. That is, the VAP is a virtual transmitter that is located on the opposite side of the reflector at the same distance from the reflector to the physical transmitter (e.g., base station).
[0068] In some embodiments, VAP 304 may facilitate the generation of non-line-of-sight (NLoS) multipath components caused by reflections from reflective surface 308. NLoS multipath components may be observed by a receiver (e.g., UE 306). Multipath components may facilitate positioning of the UE. As an example, the time of flight (ToF) of the NLoS path may correspond to the time of flight of the RF wave (e.g., positioning signal) from the virtual access point to the UE. Another example of a multipath component is the signal power (p) of the RF wave (e.g., positioning signal) transmitted from the physical transmitter 302 to the UE 306.
[0069] In some embodiments, the wireless channel may be measured to calculate the aforementioned multipath components (e.g., ToF and / or signal power), and the aforementioned multipath components may be calculated or extracted from the measured wireless channel. In some specific implementations, the UE 306 may be configured to measure ToF and / or signal power (and other wireless measurements), and track the VAP location over time based at least on those measurements, including when the UE is mobile. In some specific implementations, the physical transmitter 302 may be configured to measure ToF and / or signal power (and other wireless measurements).
[0070] like Figure 3AAs shown, the NLoS path between the physical transmitter 302 and the UE 306 and the distance between the VAP 304 and the UE 306 can be described by the total distance d1+d2. As long as the relative positions of the physical transmitter 302 and the VAP 306 do not change, d1 is also the distance of the path that the RF wave travels from the physical transmitter 302 to the reflecting surface 308. Since the VAP 304 is a mirror image of the physical transmitter 302, the path between the physical transmitter 302 and the reflecting surface 308 is also the same, and the distance between the VAP 304 and the reflecting surface 308 is also d1.
[0071] Although in some scenarios there may be a line of sight (LoS) path between the physical transmitter 302 and the UE 306 (e.g., where there are no physical obstructions along the path), multipath is likely to exist in various scenarios. Advantageously, using the VAP 304 to account for multipath scenarios (e.g., where signal reflections occur) provides a solution to traditional positioning methods. Some traditional positioning methods rely on at least three LoS measurements from physical transmitters (e.g., gNBs) to estimate the position of the UE; such LoS measurements may be used for trilateration or triangulation. However, in many environments, the line of sight to the gNB may be blocked, and therefore there may be insufficient LoS measurements. As a result, such traditional positioning methods may fail to determine the position of the UE.
[0072] However, in some embodiments disclosed herein, depending on the UE capabilities (e.g., being able to measure AoA), a single measurement of AoA and ToF may be used to determine the location of the VAP. In some other embodiments, the location of the VAP may be derived by intersecting consecutive circles around the UE, each circle having a radius based on the range of the ToF measurement. By measuring multiple times, the validity of the VAP location may be increased.
[0073] The VAP position may be fixed and consistent as long as the position of the physical transmitter 302 and the environment does not change. For example, the position of the VAP 304 does not change as long as the physical transmitter 302 and the reflective surface 308 remain in the same position relative to each other. However, the combination of reflections from different reflectors may generate a new VAP, and multiple reflections between the physical transmitter and the UE may produce multiple VAPs. Changing the position of the UE 306 does not change the position of the VAP 304, as described below with respect to FIG. Figure 3B discussed.
[0074] In some embodiments, data corresponding to the aforementioned wireless measurements (ToF, signal power, etc.) and data corresponding to locations in a wireless environment (e.g., measured by a wireless device at that location) may be used as training data for training a machine learning model, as will be described in more detail elsewhere below.
[0075] Figure 3B 306a, 306b, 306c, and the physical transmitter 302. In some embodiments, the physical transmitter 302 includes an access point or base station (e.g., a gNB), and the virtual transmitter 304 includes a VAP as discussed above. Figure 3B A first receiver (eg, UE 306a), a second receiver (eg, UE 306b), and a third receiver (eg, UE 306c) are also depicted. In some scenarios, the receiver (eg, UE 306c) may be moving.
[0076] Since the VAP 304 is a mirror image of the physical transmitter 302 across the reflective surface 308, the distance d1 traveled by the RF wave sent from the physical transmitter 302 to the reflective surface 308 is the same as the corresponding distance d1 between the VAP 304 and the reflective surface 308. Therefore, the NLoS path between the physical transmitter 302 and the UE 306b and the distance between the VAP 304 and the UE 306b can be described by the total distance d1+d2. Similarly, the distance between the VAP 304 and the UE 306a can be described by the total distance d3+d4. Similarly, the distance between the VAP 304 and the UE 306c at position A can be described by the total distance d5+d6. As the mobile UE 306c moves to another position (e.g., position B or C), the total distance can change (e.g., increase as the UE 306c moves further away from the physical transmitter 302). It can be seen that the reflection point changes as the mobile UE 306c moves. For example, when the mobile UE 306c moves from position A to position B, the corresponding reflection point is at point B instead of point A. When the mobile UE 306c moves from position B to position C, the corresponding reflection point is at point C instead of point B. It is worth noting that the location of the VAP 304 does not change when the mobile UE 306c moves, as long as the location of the physical transmitter 302 and the environment (e.g., the reflective surface 308) does not change. However, additional complexity in the environment may introduce additional VAPs, such as with respect to Figure 3C discussed.
[0077] Figure 3C304a, 304b, and UE 306 according to some embodiments. In this scenario, the physical transmitter 302 is located close to at least two reflection surfaces 308a, 308b. When the UE 306 is at position A, the RF wave may travel from the physical transmitter 302 along the reflection surface 308a to the reflection point A1, along the reflection surface 308b to the reflection point A2, and then travel to the UE 306. The reflection point A1 may exist between the first virtual transmitter (e.g., the first VAP 304a) and the reflection point A2. The reflection point A2 may exist between the second virtual transmitter (e.g., the second VAP 304b) and the UE 306.
[0078] In some cases, the UE 306 may move to positions B and / or C. When the UE 306 is at position B, the RF wave may travel from the physical transmitter 302 along the reflection surface 308a to the reflection point B1, along the reflection surface 308b to the reflection point B2, and then to the UE 306. The reflection point B1 may exist between the first VAP 304a and the reflection point B2. The reflection point B2 may exist between the second VAP 304b and the UE 306. Similarly, the RF wave may travel from the physical transmitter 302 to the reflection point C1, to the reflection point C2, to the UE 306. Although the UE may be associated with different positions in the environment, as long as the positions of the physical transmitter 302 and the environment (e.g., the reflection surfaces 308a, 308b) do not change, the positions of the VAPs 304a, 304b do not change.
[0079] In some embodiments, ToF data may be obtained based on a measurement of the time taken by a physical transmitter (e.g., a base station, an access point) to transmit a signal (e.g., a positioning signal) to a UE. In some implementations, the NLoS distance between the physical transmitter and the UE may be determined via the total distance between the corresponding VAP and the UE. Examples of the total distance may include, for example, Figure 3A or Figure 3B The discussion of d1+d2 or Figure 3B d5+d6 discussed. Therefore, ToF can be determined using ToF=d / c, where d=d1+d2 (which can also be described by (UE position)-(VAP position), i.e., the distance between the UE and the VAP), and c is the speed of light. In some implementations, the signal may include timestamp information (e.g., the transmission time) to enable the UE to measure ToF based on, for example, the reception time.
[0080] In some embodiments, signal power data may be obtained based on measurements of power at the UE. Signal power may vary depending on the location of the UE. Typically, the power may be smaller when the UE is farther away from the physical transmitter, or when the signal undergoes multiple reflections from a surface. In addition, when an electromagnetic wave hits a surface that is much longer than the wavelength of the electromagnetic wave, part of the energy of the wave is reflected, part of it is absorbed by the surface, and the rest is refracted by the surface. Therefore, as the wave travels farther through free space and / or becomes incident on a reflective surface (e.g., a wall) and reflected from the surface, energy losses result in lower signal power detected at the UE.
[0081] The aforementioned ToF and signal power may be examples of wireless measurements that may correspond to multipath components of a wireless channel between a physical transmitter (e.g., a base station) and a UE. In other words, wireless measurements (e.g., ToF and signal power) may be used to take into account reflections and obstructions in the NLoS path (multipath) between the base station and the UE, thereby enabling determination of the location information of the UE in a multipath environment. The LoS path between the base station and the UE may also be the basis for determining the location of the UE. In some implementations, examples of wireless measurements that may be used may include signal power gain, delay spread (a measure of the multipath richness of a communication channel), RSRP, and / or signal-to-noise ratio (SNR). In some implementations, as noted above, the AoA of the signal may also be an example of a wireless measurement used as the basis for the UE's location.
[0082] In addition, in some implementations, TDOA assistance data about a "reference cell" (which may also be referred to as a "reference resource") and one or more "neighboring cells" or "adjacent cells" (which may also be referred to as "target cells" or "target resources") relative to the reference cell may be provided to the UE 105 by a location server (e.g., location server 160). For example, the assistance data may provide the center channel frequency of each cell, various PRS configuration parameters (e.g., N PRS 、T PRS , muting sequence, frequency hopping sequence, PRS ID, PRS bandwidth), cell global ID, PRS signal characteristics associated with directional PRS, and / or other cell-related parameters applicable to TDOA or some other positioning method. By indicating a serving cell for UE 105 in the TDOA assistance data (e.g., where a reference cell is indicated as the serving cell), PRS-based positioning by UE 105 may be facilitated.
[0083] In some embodiments, the TDOA assistance data may also include an "expected reference signal time difference (RSTD)" parameter and the uncertainty of the expected RSTD parameter, which provides information to the UE 105 about the RSTD value that the UE 105 is expected to measure between the reference cell and each neighbor cell at its current location. The expected RSTD and the associated uncertainty may define a search window for the UE 105 within which the UE 105 is expected to measure the RSTD value. The TDOA assistance information may also include PRS configuration information parameters that allow the UE 105 to determine when a PRS positioning opportunity occurs on a signal received from each neighbor cell relative to a PRS positioning opportunity for the reference cell, and to determine a PRS sequence sent from each cell in order to measure signal ToA or RSTD.
[0084] Using RSTD measurements, known absolute or relative transmit timing for each cell, and known positions of physical transmit antennas of wireless nodes for reference and neighboring cells, UE position may be calculated (e.g., by UE 105 or by location server 160). More specifically, the RSTD of neighbor cell "k" relative to reference cell "Ref" may be given as (ToA k –ToA Ref ), where the ToA value may be measured modulo one subframe duration (1 ms) to remove the effect of measuring different subframes at different times. The ToA measurements of different cells may then be converted into RSTD measurements and transmitted by the UE 105 to the location server 160. Using (i) RSTD measurements, (ii) known absolute or relative transmit timing for each cell, (iii) known positions of physical transmit antennas for the reference cell and neighboring cells, and / or (iv) directional PRS characteristics (such as transmit direction), the UE 105 position may be determined.
[0085] Machine Learning Models
[0086] Figure 4is a block diagram illustrating a collection of wireless channel measurement data according to some embodiments. In some embodiments, one or more wireless channel measurements (generally referred to herein as wireless measurements) may be obtained based on measuring the known location of the wireless device. More specifically, one or more wireless measurements may take into account, estimate, or otherwise indicate multipath, where a signal (e.g., a positioning signal) sent by a physical transmitter (e.g., an access point or base station, such as a gNB) may traverse multiple routes when it interacts with a reflective surface (wall, ground, object, etc.). Therefore, some wireless measurements may include or be referred to as multipath components. Some examples of wireless measurements and multipath components include ToF and signal power. However, it should be understood that other types of wireless measurements (e.g., delay spread, RSRP, SNR) as discussed elsewhere herein may also be collected and used to estimate multipath.
[0087] refer to Figure 4 In some embodiments, the first set 402a of measurement data may include location data 404a and wireless measurement data 406a. In some implementations, the wireless measurement data 406a may include one or more multipath components, such as ToF and / or power measured by the wireless device. In some implementations, the location data 404a may correspond to a measured location of the wireless device. In other words, the location is a location known to be true, also referred to as a "true value" location. In some implementations, the wireless device may be a mobile device (e.g., a UE or another wireless-enabled device) having the following capabilities: (i) measuring or estimating received wireless signals, and / or (ii) determining its (wireless device's) location based on, for example, positioning techniques such as the UE-based, UE-assisted, network-based, or GNSS-based positioning methods described above. In some cases, the location of the wireless device may be obtained from any other device (e.g., a network node, a sensor, another wireless device (such as another UE)). In some cases, the wireless device may have the ability to (iii) determine the AoA of the received signal.
[0088] In some embodiments, the measuring wireless device may obtain different known real-valued positions included in, for example, position data 404a, 404b, 404n. In some embodiments, the position may be described using a coordinate system (x, y, z), a Cartesian coordinate system, a Euclidean coordinate system, a polar coordinate system, a spherical coordinate system, or other coordinate systems. In some embodiments, the position may be described relative to another wireless device or network node. For example, the wireless device may establish one or more side links with one or more anchor UEs with known positions, and may use distance and AoA and / or AoD to determine the position relative to the anchor UE. In some cases, the absolute position may be determined based on the distance and angle to the anchor UE.
[0089] Return to Figure 4 , a set of data similar to the first set 402a of measurement data may be obtained in a manner similar to the above (such as the second set 402b of measurement data and additional sets of measurement data up to or including the third set 402n of measurement data). Each of the sets 402b and 402n of measurement data may include corresponding location data (e.g., 404b, 404n) and wireless measurement data (e.g., 406b, 406n). In some embodiments, the database 410 may include sets of measurement data. The database 410 may be a data structure configured to store the aforementioned location data and wireless measurement data in a predefined format (basic array, comma separated values (CSV), text file, JavaScript object notation (JSON), Python (PY) file, etc.). In some specific implementations, the database 410 may be stored on a network side such as an access point or base station (e.g., gNB), or a server entity (e.g., location server 160, external client 180, LMF of location server 160)). In some implementations, database 410 may be stored at the wireless device and may be sent to or accessed by the network side. In various applications, database 410 may be used by at least one of such network entities to, for example, train machine learning.
[0090] Figure 5 5 is a block diagram illustrating a mechanism for training a machine learning model according to some embodiments. In some embodiments, the training of the machine learning model may be performed by a training module 500. The training module 500 may include a neural network 502. According to some implementations, the neural network 502 may include an input layer 504, an output layer 508, and one or more intermediate "hidden" layers 506a, 506b between the input layer and the output layer. In some implementations, there may be no hidden layer between the input layer and the output layer.
[0091] Neural network 502 may represent an algorithm represented by layers. Each layer may include one or more nodes, each of which may contain a value or represent a computational function having one or more weighted input connections, a transformation function that combines the inputs in some way, and / or one or more output connections (which may in turn be input connections to other nodes). Input layer 504 may be configured to receive external data. External data may be from a database (e.g., Figure 4In some implementations, a portion (e.g., 20%) of the training data may be randomly selected to be used as part of a validation set for the machine learning model. Each of the hidden layers 506a, 506b may be configured to perform at least a transformation on the input. The output layer 508 may be configured to produce a result of the transformation. In some implementations, the result may include predicted wireless measurement data, such as a signal power associated with a wireless device at a given location, a ToF at a given location. In some implementations, the result may include a predicted location of one or more VAPs. In some implementations, other types of wireless measurement data (e.g., delay spread, RSRP, SNR) may be obtained.
[0092] As an example, one or more nodes of the input layer 504 may receive location data. For example, the input node 504 may receive and / or store a coordinate or a component of a coordinate (e.g., x, y, or z). One or more hidden layers 506 may receive portions of the location data, apply one or more weights associated with a given connection, and generate a training output including a predicted signal power. There may be a correlation between the location data and the predicted signal power. The training output may include any other wireless measurement, such as ToF. The process of generating output from input may be referred to as forward propagation 510.
[0093] In some embodiments, a modeling process (e.g., linear regression) may be performed to improve the predictions made by the machine learning model. In some embodiments, the modeling process may be a logistic regression, which may determine the probability that the result in the case of a given input can be used to classify the output (e.g., yes or no, 1 or 0). In linear or logistic regression, the error (J) between the output data (e.g., predicted signal power, predicted ToF) and the wireless measurement of the tag (e.g., as a supervised learning process) may be determined, and the error may be minimized using optimization techniques such as gradient descent. In gradient descent, the error is sought to be reduced at each iteration step until the minimum error is reached. In some specific implementations, linearization may be performed to reduce the dimension and / or the learning rate may be set and / or adjusted. In some cases, a learning rate schedule may be set to vary the learning rate to achieve a global minimum error, without entering non-convergence due to an excessively large learning rate, or falling into a local minimum due to an excessively small learning rate. The process of updating the weights of the connections in the neural network 502 based on the optimization process may be referred to as back propagation 520.
[0094] Then, the forward propagation 510 can be performed again with the updated connection weights, where another backward propagation 520 is based on the updated connection weights. This cycle can be performed one or more times by the training module 500.
[0095] In some embodiments, additional input data may be utilized with the neural network 502. More specifically, a discriminator 530 and a generator 532 may optionally be implemented with the neural network 502. The discriminator is a type of neural network that is configured to learn to distinguish fake data from realistic fake data that may have the same characteristics as the training data and is generated by the generator 532. The discriminator 530 and the generator 532 may compete with each other, and the discriminator 530 may penalize the generator 532 for generating data that is easily identified as unconvincing. By using the discriminator 530 and the generator 532 together in this manner in a generative adversarial network (GAN) 534, more realistic and plausible examples may be generated by the generator 532 over time. In this manner, in some embodiments, in addition to, for example, examples generated by, for example, Figure 4 Data other than that collected by the wireless devices discussed may be used for training.
[0096] In some embodiments, the resulting output may include values (e.g., dB or dBm of signal power) included in a predefined format (such as those noted elsewhere above). In some embodiments, the resulting output may include heat map data indicating estimated or predicted wireless measurements for, for example, two-dimensional (e.g., both of x, y, or z) or three-dimensional (e.g., x, y, z) locations within an RF environment associated with a wireless network. Such heat map data may represent a collection of predictions at various locations and may include an indication of one or more predicted wireless measurements for corresponding locations of an RF environment associated with a wireless network.
[0097] In some embodiments, training of the machine learning model may be performed at a network entity such as an access point or base station (e.g., a gNB), or a server entity (e.g., location server 160, external client 180, LMF of location server 160).
[0098] In some embodiments, inference based on the trained machine learning model (e.g., predicted output of wireless measurements) may be performed at one or more network entities. In some implementations, the resulting output generated by the network entity may be transmitted to the UE. In some embodiments, inference may be performed at the UE based on the trained model received from the network entity. The resulting output may then be used as part of a positioning method to estimate the location of the UE.
[0099] Fig. 6A 6 is a block diagram illustrating inference of wireless channel measurements based on a trained machine learning model 602 according to some embodiments. In some embodiments, the trained machine learning model 602 may be configured to implement a trained neural network, such as Figure 5A neural network 502 that has undergone training and / or optimization is depicted. The trained machine learning model 602 may be configured to receive input data 604, which may include location data indicating the location of the UE 105. Based on the input data 604, the trained machine learning model 602 may generate output 606. The output may include predicted wireless measurements 606. In some embodiments, the wireless measurements may include signal power and / or ToF. In some embodiments, the wireless measurements may include the location of the VAP.
[0100] Figure 6B 6 is another block diagram illustrating inference of wireless channel measurements based on a trained machine learning model 612 according to some embodiments. In some embodiments, the trained machine learning model 612 may be configured to implement one or more trained neural networks, such as Figure 5 A neural network 502 is depicted that has undergone training and / or optimization.
[0101] In some embodiments, the trained machine learning model 612 may include at least one classifier 616a and at least one regressor 618a. As used herein, a classifier may refer to an algorithm or module configured to classify data into one or more sets of classes (e.g., yes or no, 1 or 0). As used herein, a regressor may refer to an algorithm or module configured to predict a continuous value (e.g., a wireless measurement).
[0102] Each classifier and each regressor may be associated with a corresponding VAP 620. In some cases, the presence of a VAP 620 may be determined based on a prediction or inference of the model 612 based on the input data 614. In some cases, the presence of a VAP 620 may be determined based on knowledge of the environment (e.g., the presence of reflective surfaces in the environment).
[0103] In some embodiments, the classifier may have been trained (e.g., based on Figure 5 302 to the UE location. In some embodiments, the classifier may have been trained to predict the observability of the physical transmitter 302 to the UE location. In this case, the physical transmitter observability may refer to the existence of a LoS signaling path to the UE. In other embodiments, the classifier may be used to predict the observability of other anchors. In some implementations, observability may be classified as 1 / 0 or yes / no, etc.
[0104] In some embodiments, the regressor may have been trained (e.g., according to Figure 5 The linear regression model described in ) is used to predict wireless channel measurements such as ToF and / or signal power.
[0105] In some embodiments, each classifier and each regressor can be an instance of a single trained machine learning model. In some embodiments, each classifier and each regressor can be a separately trained machine learning model. Based on the known existence of VAP 620, each instance can be generated, or each model can be trained.
[0106] In some embodiments, the trained machine learning model 612 may be configured to receive input data 614. In some embodiments, the input data 614 may include a plurality of location data. The location data may be received at one or more classifiers 616a, 616n of the corresponding VAP 620a, 620n.
[0107] In some embodiments, for each VAP, a classifier may infer the observability of the VAP to the UE location. For each UE location, a traffic access point label may represent the observable VAP to the UE location. Thus, for each VAP, a set of data may be generated, e.g., {UE location, 1 / 0}, where 1 / 0 indicates whether the VAP or UE is observable (1) or unobservable (0). In some embodiments, for each VAP, a regressor may predict the signal power generated by the traffic access point at the UE location. For each UE location, the assigned traffic access point to the VAP may generate data as, e.g., {UE location, signal power} for the VAP.
[0108] In some implementations, wireless channel measurements for the UE location may be predicted only if observability is positive (eg, 1).A set of wireless measurement data 622 may be obtained from available VAPs in the environment.
[0109] Figure 6C 618 is a block diagram illustrating the use of a classifier 616 and a regressor 618 in a machine learning model to predict and obtain wireless measurements according to some embodiments. As discussed above, if the classifier 616 determines that the VAP is observable ("1") at the UE location (given by the received input data 614), the regressor 618 can use the input data 614 to infer wireless measurements 622n at the UE location. Examples of wireless measurements may include ToF, signal power, and other wireless measurements as discussed elsewhere herein. If the VAP or UE is not observable ("0"), no wireless measurement is predicted, estimated, or otherwise determined.
[0110] method
[0111] Figure 7 is a flow chart of a method 700 of predicting wireless channel data associated with a user equipment (UE) of a wireless network according to some embodiments. Figure 7The functional components illustrated in one or more of the illustrated blocks may be performed by hardware (e.g., a processor) and / or software components of a UE, a base station, a server, or a computer-readable device, the computer-readable device including a storage medium storing computer-readable and / or computer-executable instructions that are configured to cause the at least one processor device or another computerized device to perform operations when executed by the processor device. Fig.10 , Fig.11 and Fig.12 , example components of a UE, a base station, and a server are respectively illustrated in FIG. 4 , and these example components are described in more detail below.
[0112] It should also be noted that the operations of method 700 may be performed in any suitable order, and need not be Figure 7 In addition, method 700 may include comparing Figure 7 Those depicted may require more or fewer operations to perform the prediction.
[0113] At block 710, method 700 may include obtaining the location of the UE. In some embodiments, at least one positioning technique such as a UE-based, UE-assisted, network-based, or GNSS-based positioning method may be used to determine the location of the UE. In some embodiments, the location may already be known based on previous measurements made by the UE (e.g., using a UE-based method) or by the network (e.g., using a UE-assisted method). The known location may be stored on the network side at a network entity (e.g., a base station, a server, or a LMF), or provided to the UE.
[0114] At block 720, method 700 may include using the machine learning model to generate outputs including one or more predicted wireless measurements given the location of the UE. In some embodiments, the machine learning model may be obtained by the UE from a network entity and implemented by the UE to generate the outputs at the UE. In some embodiments, the machine learning model may be implemented by the network entity to generate the outputs at the network entity.
[0115] In some embodiments, the machine learning model may be trained (step 722) by at least obtaining a training data set (box 723) and performing an optimization with respect to the multipath component data and the true value locations (box 724).
[0116] In some implementations, the obtained training data set may include (i) multipath component data and (ii) the true value position of the wireless device. The wireless device may be configured to measure the multipath component and may be separate from the UE. More directly, the wireless device may be a device for collecting training data, and the training data may include the multipath component data and the true value position of the wireless device.
[0117] In some implementations, the multipath component data may correspond to a virtual location associated with an access point of a wireless network (e.g., a gNB or other base station configured for 3GPP-based data communications). The virtual location may correspond to a virtual access point (VAP) and may include a location determined based on the location of the access point. More specifically, the VAP may be at a location that is mirrored relative to the location of the access point across a reflective surface (e.g., a wall), such as (for example) FIG. 3A to FIG. 3C Thus, a VAP may be associated with one or more known locations.
[0118] In some cases, the multipath component includes a wireless measurement that may indicate multipath in the NLoS signal between a transmitter (e.g., a base station) and a receiver (e.g., a UE) and may include a time of flight (ToF) between a VAP and a wireless device. The multipath component may also include signal power (p). Other examples of multipath components may include signal power gain, delay spread, RSRP, or SNR. One of ordinary skill in the relevant art will recognize other parameters and measurements that may be used.
[0119] In some embodiments, optimization may include gradient descent of the regression model, which may involve minimizing the error J, such as with respect to Figure 5 As discussed. In some implementations, the regression model may generate a training output that may include a predicted multipath component. Optimization may include iterative minimization of an error between at least the predicted multipath component and the labeled multipath component. Optimization may include updating weights associated with node connections via forward propagation and back propagation. In some cases, the regression model may be a logistic regression model that may be used to train a classifier such as classifier 616 and predict a classification, for example, the observability of a VAP. In some cases, the regression model may be a linear regression model that may be used to train a regressor such as regressor 618 and output a value of a wireless measurement.
[0120] In various specific implementations, potential applications of the output may include (1) estimating or predicting a wireless channel or a multipath component of a wireless channel (e.g., delay spread, RSRP, SNR); (2) predictive scheduling decisions, such as rank selection, allocation of resource blocks (RBs, which are blocks of OFDM symbols and subcarriers), coordinated multipoint (CoMP) modes (CoMP may refer to the use of multiple base stations to serve one UE, and may have different modes and multiple sets of base stations that may be selected to serve the UE; example CoMP modes include coherent and non-coherent joint transmission), and set decision making (selecting which subset of base stations and the mode of operation based on channel conditions); (3) applications in ultra-reliable low-latency communications (URLLC), which depend on a priori information about the channel; or (4) dynamic adaptation of communication parameters (such as reference signal (RS) patterns, periodicity of channel state feedback (CSF, which may contain a metric indicating channel quality fed back to the base station), cyclic prefix (CP) length, etc.).
[0121] Optionally, at block 730, method 700 may include providing an output to the UE. As mentioned above, in some embodiments, a machine learning model may have been implemented by a network entity to output a predicted wireless measurement. The output may be used by the network entity to estimate the wireless channel or multipath components of the wireless channel, or by the UE to perform the same when provided to the UE.
[0122] Optionally, at block 740, method 700 may include determining whether to update the location of the VAP, or whether to provide at least one of the one or more wireless measurements to a server entity for use in positioning the UE. In some embodiments, a new estimated location of the VAP may be determined based on a training data set (e.g., whether the location of the VAP is within an expected threshold range) or a heat map generated by a machine learning model. In some specific implementations, the location of the VAP may also be used to determine or estimate the location of the physical transmitter as a mirrored location of the VAP. If the determined location of the VAP does not need to be updated, the measurement may be sent to a server entity (e.g., location server 160, external client 180, LMF) for use in positioning the UE (e.g., UE-assisted positioning).
[0123] Optionally, at block 742, method 700 may include updating the location of the virtual access point based on the determination at block 740. Optionally, at block 744, method 700 may include providing at least one of the one or more wireless measurements to a server for use in positioning of the UE based on the determination at block 740. In some cases, both of the foregoing may occur.
[0124] Figure 8 is a flowchart of a method 800 for training a machine learning model according to some embodiments. Figure 8 The functional components illustrated in one or more of the illustrated blocks may be performed by hardware (e.g., a processor) and / or software components of a base station, a server, or a computer-readable device, which includes a storage medium storing computer-readable and / or computer-executable instructions that are configured to cause the at least one processor device or another computerized device to perform operations when executed by the processor device. Fig.11 and Fig.12 , example components of a base station and a server are respectively illustrated in FIG. 4 , and these example components are described in more detail below.
[0125] It should also be noted that the operations of method 800 may be performed in any suitable order, and need not be Figure 8 In addition, method 800 may include comparing Figure 8 Those depicted may require more or fewer operations to perform training.
[0126] At box 810, method 800 includes obtaining multiple training data. In some embodiments, the multiple training data may include one or more of the following: (i) the location of the wireless device within the wireless network, (ii) one or more multipath components associated with at least one VAP (as discussed elsewhere herein), or (iii) the observability of the virtual access point at the location of the wireless device. A combination of the foregoing may be included with the multiple training data. In some embodiments, the wireless device may be a mobile device (e.g., a UE or another wireless-enabled device) having the following capabilities: (i) measuring or estimating received wireless signals, (ii) determining its (wireless device's) location, and / or (iii) determining the AoA of the received signal. In some specific implementations, the location may be a true value location. In some specific implementations, the location may be obtained from another entity other than the wireless device, such as a network entity. Note that the wireless device may be separate from the UE that will use the model to infer information (e.g., wireless measurements such as multipath components) after training.
[0127] At block 820, method 800 includes optimizing the machine learning model. In some embodiments, the optimization may be performed using the obtained plurality of training data to obtain a trained machine learning model configured to generate wireless parameters related to the location of the user equipment (UE). More specifically, in some implementations, the optimization may include a gradient descent of a regression model, such as with respect to Figure 5 In some implementations, the optimization can include updating weights associated with node connections via forward propagation and back propagation.
[0128] In some embodiments, the machine learning model may be a classifier (logistic regression model) configured to predict the observability of the UE to the VAP. In some embodiments, the machine learning model may be a regressor (linear regression model) configured to predict one or more wireless measurements (e.g., ToF or signal power).
[0129] Fig. 9 is a flow chart of a method 900 of predicting wireless channel data associated with a user equipment (UE) of a wireless network according to some embodiments. Fig. 9 The functional components illustrated in one or more of the illustrated blocks may be performed by hardware (e.g., a processor) and / or software components of a base station, a server, or a computer-readable device, which includes a storage medium storing computer-readable and / or computer-executable instructions that are configured to cause the at least one processor device or another computerized device to perform operations when executed by the processor device. Fig.11 and Fig.12 , example components of a base station and a server are respectively illustrated in FIG. 4 , and these example components are described in more detail below.
[0130] It should also be noted that the operations of method 900 may be performed in any suitable order, and need not be Fig. 9 In addition, method 800 may include comparing Fig. 9 Those depicted may require more or fewer operations to perform the prediction.
[0131] At block 910, method 900 may include training a first machine learning model. In some embodiments, the first machine learning model may include a classifier, such as a Figure 6B and Figure 6C Thus, the first machine learning model may be configured to predict the observability of a VAP to a UE location.
[0132] In some embodiments, training may be based at least on one or more locations of the wireless device and one or more ground truth observables corresponding to one or more locations of the wireless device relative to the virtual access point. In some implementations, the wireless device is a device separate from the UE and may be a mobile device (e.g., a UE or another wireless-enabled device) capable of (i) measuring or estimating received wireless signals, (ii) determining its (wireless device's) location, and / or (iii) determining the AoA of received signals. In some implementations, the location may be a ground truth location. In some implementations, the location may be obtained from another entity other than the wireless device, such as a network entity.
[0133] At block 920, method 900 may include obtaining a training data set. In some embodiments, the training data set may include a set of training multipath components associated with a VAP having a defined relationship with a location of an access point of the wireless network, each set of training multipath components including one or more training wireless measurements associated with the VAP. The training data set may also include a plurality of ground truth locations associated with the set of multipath components.
[0134] In other words, each VAP may have its own set of multipath components. A given set of multipath components may include one or more of ToF, signal power, or other related wireless measurements. Each VAP may also be associated with one or more known locations. Each known location may be associated with, for example, ToF and signal power. For example, multiple locations may enable the generation of a heat map of ToF and / or signal power relative to location. Therefore, in some specific implementations, each VAP may be associated with, for example, one or more known locations, one or more ToF measurements, and one or more signal power measurements, which are performed by, for example, a wireless device.
[0135] At block 930, method 900 may include training a second machine learning model based at least on the obtained training data set. In some embodiments, the first machine learning model may include a regressor, such as Figure 6B and Figure 6C Described regressor 618. In some embodiments, training may be performed according to training methods described elsewhere herein (eg, implementing block 722, method 800, or training module 500).
[0136] At box 940, method 900 may include implementing the trained second machine learning model within the wireless network to predict wireless channel data. In some embodiments, the implementation of the trained second machine learning model may include providing the trained second machine learning model to the UE when the training has been performed at a network entity (e.g., a base station, a server, a LMF). In some embodiments, the implementation of the trained second machine learning model may include generating a prediction of the wireless channel data. In some embodiments, the prediction of the wireless channel data may include an inference of the wireless channel data. The wireless channel data may include wireless measurements such as ToF, signal power, etc. In some embodiments, the wireless channel data may be predicted in response to the observability of the VAP or UE location being positive ("1"), as determined by the trained first machine learning model.
[0137] Device
[0138] Fig.10 is a block diagram of an implementation of UE 105, which may be described herein above (e.g., in conjunction with Figure 7) is utilized as described in . For example, UE 105 may execute Figure 7 However, the UE may also be configured to perform one or more of the functions of the method shown. Figure 8 and Fig. 9 One or more functions of the method shown, such as obtaining training data (e.g., block 820 or 920) and / or predicting wireless channel data (e.g., block 940). It should be noted that Fig.10 It is intended only to provide a generalized illustration of various components, any or all of which may be utilized as appropriate. It may be noted that in some instances, Fig.10 The illustrated components may be localized into a single physical device and / or distributed among various networked devices that may be located at different physical locations. In addition, as previously mentioned, the functionality of the UE discussed in the previously described embodiments may be provided by Fig.10 One or more of the illustrated hardware and / or software components are implemented.
[0139] UE 105 is shown as including hardware elements that may be electrically coupled via bus 1005 (or may communicate in other ways as appropriate). The hardware elements may include a processor 1010, which may include, but is not limited to, one or more general-purpose processors (e.g., application processors), one or more special-purpose processors (such as digital signal processor (DSP) chips, graphics acceleration processors, application-specific integrated circuits (ASICs), etc.), and / or other processing structures or components. Processor 1010 may include one or more processing units, which may be housed in a single integrated circuit (IC) or multiple ICs. Fig.10 As shown, some embodiments may have a separate DSP 1020, depending on the desired functionality. Position determination and / or other determinations based on wireless communication (discussed below) may be provided in the processor 1010 and / or the wireless communication interface 1030. The UE 105 may also include: one or more input devices 1070, which may include but are not limited to one or more keyboards, touch screens, touch pads, microphones, buttons, dials, switches, etc.; and one or more output devices 1015, which may include but are not limited to one or more displays (e.g., touch screens), light emitting diodes (LEDs), speakers, etc.
[0140] UE 105 may also include a wireless communication interface 1030, which may include but is not limited to a modem, a network card, an infrared communication device, a wireless communication device and / or a chipset (such as Device, IEEE 802.11 device, IEEE 802.15.4 device, Wi-Fi device, WiMAX device, WAN device and / or various cellular devices, etc.), which can enable UE 105 to communicate with other devices as described in the above embodiments. The wireless communication interface 1030 can allow the use of the TRP of the network, for example, via eNB, gNB, ng-eNB, access point, various base stations and / or other access node types, and / or other network components, computer systems and / or any other electronic device communicatively coupled to the TRP to communicate (e.g., send and receive) data and signaling, as described herein. Communication can be performed via one or more wireless communication antennas 1032 that transmit and / or receive wireless signals 1034. According to some embodiments, the wireless communication antenna 1032 may include multiple discrete antennas, antenna arrays, or any combination thereof. Antenna 1032 may be able to use beams (e.g., Tx beams and Rx beams) to send and receive wireless signals. Beamforming can be performed using digital and / or analog beamforming techniques using corresponding digital and / or analog circuits. Wireless communication interface 1030 may include such circuitry.
[0141] Depending on the desired functionality, the wireless communication interface 1030 may include separate receivers and transmitters, or any combination of transceivers, transmitters, and / or receivers to communicate with base stations (e.g., ng-eNBs and gNBs) and other terrestrial transceivers such as wireless devices and access points. The UE 105 may communicate with different data networks, which may include various network types. For example, a wireless wide area network (WWAN) may be a CDMA network, a time division multiple access (TDMA) network, a frequency division multiple access (FDMA) network, an orthogonal frequency division multiple access (OFDMA) network, a single carrier frequency division multiple access (SC-FDMA) network, a WiMAX (IEEE 802.16) network, and the like. A CDMA network may implement one or more RATs, such as WCDMA, etc. Includes IS-95, IS-2000, and / or IS-856 standards. A TDMA network may implement GSM, Digital Advanced Mobile Phone System (D-AMPS), or some other RAT. An OFDMA network may employ LTE, Advanced LTE, 5G NR, and the like. 5G NR, LTE, Advanced LTE, GSM, and WCDMA are described in documents from 3GPP. Described in documents from an organization named "3rd Generation Partnership Project 2" (3GPP2). 3GPP and 3GPP2 documents are publicly available. A wireless local area network (WLAN) may also be an IEEE 802.11x network, while a wireless personal area network (WPAN) may be a Bluetooth network, IEEE 802.15x, or some other type of network. The techniques described herein may also be used for any combination of WWAN, WLAN, and / or WPAN.
[0142] The UE 105 may also include sensors 1040. The sensors 1040 may include, but are not limited to, one or more inertial sensors and / or other sensors (e.g., accelerometers, gyroscopes, cameras, magnetometers, altimeters, microphones, proximity sensors, light sensors, barometers, etc.), some of which may be used to obtain location-related measurements and / or other information.
[0143] Embodiments of UE 105 may also include a global navigation satellite system (GNSS) receiver 1080 that is capable of receiving signals 1084 from one or more GNSS satellites using antenna 1082 (which may be the same as antenna 1032). Positioning based on GNSS signal measurements may be used to supplement and / or incorporate the techniques described herein. GNSS receiver 1080 may extract the location of UE 105 from GNSS satellites 110 of GNSS systems such as Global Positioning System (GPS), Galileo, GLONASS, Quasi-Zenith Satellite System (QZSS) over Japan, IRNSS over India, BeiDou Navigation Satellite System (BDS) over China, etc. using conventional techniques. In addition, the GNSS receiver 1080 may be used for various augmentation systems (e.g., satellite-based augmentation systems (SBAS)) that may be associated with or otherwise enabled for use with one or more global and / or regional navigation satellite systems, such as, for example, the Wide Area Augmentation System (WAAS), the European Geostationary Navigation Overlay Service (EGNOS), the Multipurpose Satellite Augmentation System (MSAS), and the Geographic Augmented Navigation System (GAGAN), among others.
[0144] It can be noted that although Fig.10GNSS receiver 1080 is illustrated as a different component in the examples, but the embodiments are not limited to this. As used herein, the term "GNSS receiver" may include hardware and / or software components configured to obtain GNSS measurements (measurements from GNSS satellites). Therefore, in some embodiments, the GNSS receiver may include a measurement engine executed (as software) by one or more processors (such as processor 1010, DSP 1020 and / or a processor within wireless communication interface 1030 (e.g., in a modem)). The GNSS receiver may also optionally include a positioning engine that can use the GNSS measurements from the measurement engine to determine the position of the GNSS receiver using an extended Kalman filter (EKF), a weighted least squares method (WLS), a hatch filter, a particle filter, etc. The positioning engine may also be executed by one or more processors (such as processor 1010 or DSP 1020).
[0145] The UE 105 may also include and / or communicate with memory 1060. The memory 1060 may include, but is not limited to, local and / or network accessible storage, disk drives, drive arrays, optical storage devices, solid-state storage devices (such as random access memory (RAM) and / or read-only memory (ROM)), which may be programmable, flash-updatable, etc. Such storage devices may be configured to implement any suitable data storage, including, but not limited to, various file systems, database structures, etc.
[0146] The memory 1060 of the UE 105 may also include software elements ( Fig.10 105 ), these software elements include operating systems, device drivers, executable libraries and / or other code (such as one or more applications), which may include computer programs provided by various embodiments and / or may be designed to implement methods provided by other embodiments and / or configure systems provided by other embodiments, as described herein. By way of example only, one or more procedures described with respect to the methods discussed above may be implemented as code and / or instructions in memory 1060 that can be executed by UE 105 (and / or processor 1010 or DSP 1020 within UE 105). In some embodiments, such code and / or instructions may then be used to configure and / or adapt a general purpose computer (or other device) to perform one or more operations according to the described methods.
[0147] Fig.11 is a block diagram of an embodiment of a base station 120, which may be described herein above (e.g., in conjunction with Figures 7 to 9 ) is used as described. It should be noted that Fig.11It is intended only to provide a generalized illustration of the various components, any or all of which may be utilized as appropriate. In some embodiments, base station 120 may correspond to a gNB, ng-eNB, and / or (more generally) a TRP.
[0148] Base station 120 is shown as including hardware elements that may be electrically coupled via bus 1105 (or may communicate in other ways as appropriate). The hardware elements may include processor 1110, which may include, but is not limited to, one or more general-purpose processors, one or more special-purpose processors (such as DSP chips, graphics acceleration processors, ASICs, etc.), and / or other processing structures or components. Fig.11 As shown, some embodiments may have a separate DSP 1120, depending on the desired functionality. According to some embodiments, wireless communication-based position determination and / or other determinations may be provided in the processor 1110 and / or the wireless communication interface 1130 (discussed below). The base station 120 may also include one or more input devices, which may include but are not limited to a keyboard, a display, a mouse, a microphone, buttons, a dial, a switch, etc.; and one or more output devices, which may include but are not limited to a display, a light emitting diode (LED), a speaker, etc.
[0149] The base station 120 may also include a wireless communication interface 1130, which may include but is not limited to a modem, a network card, an infrared communication device, a wireless communication device and / or a chipset (such as Devices, IEEE 802.11 devices, IEEE 802.15.4 devices, Wi-Fi devices, WiMAX devices, cellular communication facilities, etc.), etc., which can enable the base station 120 to communicate as described herein. The wireless communication interface 1130 can permit the communication (e.g., sending and receiving) of data and signaling to UEs, other base stations / TRPs (e.g., eNBs, gNBs, and ng-eNBs), and / or other network components, computer systems, and / or any other electronic devices described herein. Communication can be performed via one or more wireless communication antennas 1132 that transmit and / or receive wireless signals 1134.
[0150] The base station 120 may also include a network interface 1180, which may include support for wired communication technologies. The network interface 1180 may include a modem, a network card, a chipset, etc. The network interface 1180 may include one or more input and / or output communication interfaces to permit data to be exchanged with a network, a communication network server, a computer system, and / or any other electronic device described herein.
[0151] In many embodiments, the base station 120 may also include a memory 1160. The memory 1160 may include, but is not limited to, local and / or network accessible storage, disk drives, drive arrays, optical storage devices, solid-state storage devices such as RAM and / or ROM, which may be programmable, flash-updatable, etc. Such storage devices may be configured to implement any suitable data storage, including, but not limited to, various file systems, database structures, etc.
[0152] The memory 1160 of the base station 120 may also include software elements ( Fig.11 110), these software elements include operating systems, device drivers, executable libraries and / or other code such as one or more applications, which may include computer programs provided by various embodiments, and / or may be designed to implement methods provided by other embodiments, and / or configure systems provided by other embodiments, as described herein. By way of example only, one or more procedures described with respect to the methods discussed above may be implemented as code and / or instructions in memory 1160 that can be executed by base station 120 (and / or processor 1110 or DSP 1120 within base station 120). In some embodiments, such code and / or instructions may then be used to configure and / or adapt a general purpose computer (or other device) to perform one or more operations according to the described methods.
[0153] Fig.12 is a block diagram of an embodiment of a computer system 1200 that may be used, in whole or in part, to provide one or more network components (e.g., Figure 1 The location server 160 of the present invention may be used as described above (e.g., in conjunction with Figures 7 to 9 ) is used as described. It should be noted that Fig.12 It is intended only to provide a generalized illustration of the various components, any or all of which may be utilized as appropriate. Fig.12 The following broadly illustrates how individual system elements can be implemented in a relatively separate or relatively more integrated manner. Fig.12 The illustrated components may be localized into a single device and / or distributed among various networked devices that may be located at different geographical locations.
[0154] The computer system 1200 is shown as including hardware elements that can be electrically coupled via a bus 1205 (or can communicate in other ways as appropriate). The hardware elements may include a processor 1210, which may include but is not limited to one or more general-purpose processors, one or more special-purpose processors (such as digital signal processing chips, graphics acceleration processors, etc.) and / or other processing structures, which can be configured to perform one or more of the methods described herein. The computer system 1200 may also include one or more input devices 1215, which may include but are not limited to a mouse, a keyboard, a camera, a microphone, etc.; and one or more output devices 1220, which may include but are not limited to a display device, a printer, etc.
[0155] The computer system 1200 may also include (and / or communicate with) one or more non-transitory storage devices 1225, which may include, but are not limited to, local and / or network accessible storage, and / or may include, but are not limited to, disk drives, drive arrays, optical storage devices, solid-state storage devices such as RAM and / or ROM, which may be programmable, flash-updatable, etc. Such storage devices may be configured to implement any suitable data storage, including, but not limited to, various file systems, database structures, etc. Such data storage may include databases and / or other data structures for storing and managing messages and / or other information to be transmitted to one or more devices via the hub, as described herein.
[0156] The computer system 1200 may also include a communication subsystem 1230, which may include wireless communication technologies managed and controlled by a wireless communication interface 1233, as well as wired technologies (such as Ethernet, coaxial communication, universal serial bus (USB), etc.). The wireless communication interface 1233 may include one or more wireless transceivers, which may transmit and receive wireless signals 1255 (e.g., signals according to 5G NR or LTE) via a wireless antenna 1250. Therefore, the communication subsystem 1230 may include a modem, a network card (wireless or wired), an infrared communication device, a wireless communication device, and / or a chipset, etc., which may enable the computer system 1200 to communicate with any device on the corresponding network including user equipment (UE), base stations and / or other TRPs, and / or any other electronic device described herein on any or all of the communication networks described herein. Therefore, the communication subsystem 1230 can be used to receive and transmit data, as described in the embodiments of this article.
[0157] In many embodiments, the computer system 1200 will also include a working memory 1235, which may include a RAM or ROM device, as described above. The software elements shown as being located within the working memory 1235 may include an operating system 1240, device drivers, executable libraries, and / or other code such as one or more applications 1245, which may include computer programs provided by various embodiments and / or may be designed to implement methods provided by other embodiments and / or configure systems provided by other embodiments, as described herein. By way of example only, one or more of the procedures described for the methods discussed above may be implemented as code and / or instructions executable by a computer (and / or a processor within a computer); then, in one aspect, such code and / or instructions may be used to configure and / or adapt a general purpose computer (or other device) to perform one or more operations according to the described methods.
[0158] The set of these instructions and / or codes may be stored on a non-transitory computer-readable storage medium such as the storage device 1225 described above. In some cases, the storage medium may be incorporated into a computer system such as the computer system 1200. In other embodiments, the storage medium may be separate from the computer system (e.g., a removable medium such as an optical disk), and / or may be provided in an installation package so that the storage medium can be used to program, configure, and / or adapt a general-purpose computer having the instructions / code stored thereon. The instructions may take the form of executable code that can be executed by the computer system 1200, and / or may take the form of source code and / or installable code that, when compiled and / or installed on the computer system 1200 (e.g., using any of a variety of commonly available compilers, installers, compression / decompression utilities, etc.), then takes the form of executable code.
[0159] It will be apparent to those skilled in the art that basic variations may be made according to specific requirements. For example, customized hardware may also be used, and / or specific elements may be implemented in hardware, software (including portable software, such as applets, etc.), or both. In addition, connections to other computing devices such as network input / output devices may be employed.
[0160] With reference to the accompanying drawings, components that may include memory may include non-transient machine-readable media. As used herein, the terms "machine-readable medium" and "computer-readable medium" refer to any storage medium that participates in providing data that enables a machine to operate in a particular manner. In the embodiments provided above, various machine-readable media may be involved when providing instructions / codes to a processor and / or other devices for execution. Additionally or alternatively, machine-readable media may be used to store and / or carry such instructions / codes. In many specific implementations, computer-readable media are physical and / or tangible storage media. Such media may take many forms, including but not limited to non-volatile media and volatile media. Common forms of computer-readable media include, for example: magnetic and / or optical media, any other physical media with hole patterns, RAM, programmable ROM (PROM), erasable PROM (EPROM), FLASH-EPROM, any other memory chip or memory box, or any other medium from which a computer can read instructions and / or codes.
[0161] The methods, systems and devices discussed herein are examples. Various embodiments may appropriately omit, replace or add various procedures or components. For example, the features described with reference to certain embodiments may be combined in various other embodiments. The different aspects and elements of the embodiments may be combined in a similar manner. The various components of the accompanying drawings provided herein may be embodied in hardware and / or software. Moreover, technology may evolve, and therefore many elements are examples, which do not limit the scope of the present disclosure to those specific examples.
[0162] It has proven convenient at times, primarily for reasons of common usage, to refer to such signals as bits, information, values, elements, symbols, characters, variables, terms, numbers, digital symbols, and the like. It should be understood, however, that all of these or similar terms are to be associated with the appropriate physical quantities and are merely convenient labels. Unless specifically stated otherwise, as will be apparent from the above discussion, it should be appreciated that throughout this specification, discussions utilizing terms such as "processing," "computing," "calculating," "determining," "ascertaining," "identifying," "correlating," "measuring," "performing," and the like refer to the actions or processes of a specific apparatus, such as a special-purpose computer or similar special-purpose electronic computing device. Thus, in the context of this specification, a special-purpose computer or similar special-purpose electronic computing device is capable of manipulating or transforming signals, typically expressed as physical, electronic, electrical, or magnetic quantities in a memory, register, or other information storage device, a transmitting device, or a display device of the special-purpose computer or similar special-purpose electronic computing device.
[0163] As used herein, the terms "and" and "or" may include various meanings that are also intended to depend at least in part on the context in which such terms are used. Generally, "or", if used in association with a list, such as A, B, or C, is intended to mean A, B, and C (inclusive meaning is used here) as well as A, B, or C (exclusive meaning is used here). In addition, as used herein, the term "one or more" may be used to describe any feature, structure, or characteristic in the singular, or may be used to describe a certain combination of features, structures, or characteristics. However, it should be noted that this is merely an illustrative example, and the subject matter claimed is not limited to this example. In addition, the term "at least one of", if used in association with a list, such as A, B, or C, may be interpreted to mean any combination of A, B, and / or C, such as A, AB, AA, AAB, AABBCCC, etc.
[0164] Several embodiments have been described, and various modifications, alternative configurations, and equivalents may be used without departing from the scope of the present disclosure. For example, the above elements may be components of a larger system, where other rules may take precedence over the application of the various embodiments or otherwise modify the application of the various embodiments. Moreover, several steps may be taken before, during, or after considering the above elements. Accordingly, the above description does not limit the scope of the present disclosure.
[0165] In view of this description, various embodiments may include different combinations of features. Specific implementation examples are described in the following numbered clauses:
[0166] Item 1: A method for predicting wireless channel data associated with a user equipment (UE) of a wireless network, the method comprising: obtaining a position of the UE; and given the position of the UE, using a first machine learning model to generate an output comprising one or more predicted wireless measurements, the first machine learning model being trained by at least the following operations: obtaining a training data set comprising (i) multipath component data and (ii) a true value position of a wireless device, the multipath component data corresponding to a virtual access point associated with an access point of the wireless network, the virtual access point comprising a position determined based on the position of the access point; and performing optimization with respect to the multipath component data and the true value position.
[0167] Clause 2: A method according to clause 1, wherein: the multipath component data includes data corresponding to flight time, signal power, or a combination thereof; and the one or more predicted wireless measurements include flight time, signal power, or a combination thereof.
[0168] Clause 3: A method according to any one of clauses 1 to 2, wherein the method is performed by the UE, and the method also includes receiving the first machine learning model by the UE from a network entity of the wireless network.
[0169] Clause 4: A method according to any one of clauses 1 to 3, wherein the method is performed by a base station configured for data communications based on the Third Generation Partnership Project (3GPP), and the method also includes transmitting, by the base station, the output including the one or more predicted wireless measurements to the UE.
[0170] Clause 5: The method according to any one of clauses 1 to 4 further includes using a second machine learning model to determine the observability of the virtual access point to the UE; wherein the generation of the output of the one or more predicted wireless measurements is performed in response to the observability of the virtual access point being positive.
[0171] Clause 6: A method according to any one of clauses 1 to 5, wherein the first machine learning model comprises a regressor and the second machine learning model comprises a classifier, which is trained using a training data set comprising (i) the multipath component data, (ii) the true value location of the wireless device, and (iii) observability of the wireless device by a virtual access point.
[0172] Clause 7: A method as described in any of clauses 1 to 6, wherein the output includes a heat map, the heat map including indications of the one or more predicted wireless measurements for corresponding locations regarding a radio frequency (RF) environment associated with the wireless network.
[0173] Clause 8: The method of any of clauses 1 to 7, wherein the location of the virtual access point is at a distance from a reflective surface to the virtual access point that is equal to the distance from the reflective surface to the location of the access point.
[0174] Clause 9: A method according to any one of clauses 1 to 8, wherein the first machine learning model is further trained by generating training outputs, the training outputs comprising predicted multipath components; and the optimization comprises iterative minimization of the error between at least the predicted multipath components and the labeled multipath components.
[0175] Clause 10: The method according to any one of clauses 1 to 9 further includes: updating the location of the virtual access point based on the predicted one or more wireless measurements, or providing at least one of the one or more wireless measurements to a server for use in positioning the UE.
[0176] Clause 11: A network node within a wireless network, the network node comprising: at least one wireless communication interface; a memory; and one or more processors, the one or more processors being communicatively coupled to the at least one wireless communication interface and the memory, and the one or more processors being configured to: obtain a location of a user equipment (UE); and given the location of the UE, use a first machine learning model to generate an output comprising one or more predicted wireless measurements; wherein the first machine learning model is trained by at least the following operations: obtaining a training data set comprising (i) multipath component data and (ii) a true value location of a wireless device, the multipath component data corresponding to a virtual access point associated with an access point of the wireless network, the virtual access point comprising a location determined based on the location of the access point; and performing optimization with respect to the multipath component data and the true value location.
[0177] Clause 12: A network node according to clause 11, wherein: the multipath component data includes data corresponding to flight time, signal power, or a combination thereof; and the one or more predicted wireless measurements include flight time, signal power, or a combination thereof.
[0178] Clause 13: A network node according to any one of clauses 11 to 12, wherein the network node is configured for data communication based on the Third Generation Partnership Project (3GPP); and the one or more processors are further configured to transmit the output including the one or more predicted wireless measurements to the UE via the at least one wireless communication interface by the network node.
[0179] Clause 14: A network node according to any one of clauses 11 to 13, wherein the one or more processors are further configured to determine the observability of the virtual access point to the UE using a second machine learning model; and the generation of the output including the one or more predicted wireless measurements is performed in response to the observability of the virtual access point being positive.
[0180] Clause 15: A network node according to any one of clauses 11 to 14, wherein the first machine learning model comprises a regressor and the second machine learning model comprises a classifier, which is trained using a training data set comprising (i) the multipath component data, (ii) the true value location of the wireless device, and (iii) observability of the wireless device by a virtual access point.
[0181] Clause 16: A network node as described in any of clauses 11 to 15, wherein the output comprises a heat map including indications of the one or more predicted wireless measurements for corresponding locations regarding a radio frequency (RF) environment associated with the wireless network.
[0182] Clause 17: A network node as set forth in any one of clauses 11 to 16, wherein the location of the virtual access point is at a distance from a reflective surface to the virtual access point, the distance being equal to the distance from the reflective surface to the location of the access point.
[0183] Clause 18: A network node according to any one of clauses 11 to 17, wherein the first machine learning model is further trained by generating training outputs, the training outputs comprising predicted multipath components; and the optimization comprises iterative minimization of the error between at least the predicted multipath components and the labeled multipath components.
[0184] Clause 19: A computer-readable device comprising a storage medium comprising a plurality of instructions configured to, when executed by one or more processors, cause the computerized device to: obtain a location of a user equipment (UE) within a wireless network; and, given the location of the UE, use a first machine learning model to generate an output comprising one or more predicted wireless measurements; wherein the first machine learning model is trained by at least the following operations: obtaining a training data set comprising (i) multipath component data and (ii) a true value location of a wireless device, the multipath component data corresponding to a virtual access point associated with an access point of the wireless network, the virtual access point comprising a location determined based on the location of the access point; and performing optimization with respect to the multipath component data and the true value location.
[0185] Clause 20: A computer-readable device according to clause 19, wherein: the multipath component data includes data corresponding to flight time, signal power, or a combination thereof; and the one or more predicted wireless measurements include flight time, signal power, or a combination thereof.
[0186] Clause 21: A computer-readable device according to any one of clauses 19 to 20, wherein the computerized device includes the UE; and the plurality of instructions are further configured to cause the UE to receive the first machine learning model from a network entity of the wireless network when executed by the one or more processors.
[0187] Clause 22: A computer-readable device according to any one of clauses 19 to 21, wherein the computerized device includes a base station configured for data communication based on the Third Generation Partnership Project (3GPP); and the plurality of instructions are further configured to, when executed by the one or more processors, cause the base station to transmit the output including the one or more predicted wireless measurements to the UE.
[0188] Clause 23: A computer-readable device according to any one of clauses 19 to 22, wherein the plurality of instructions are further configured to cause the computerized device to use a second machine learning model to determine the observability of the virtual access point to the UE when executed by the one or more processors; and the generation of the output including the one or more predicted wireless measurements is performed in response to the observability of the virtual access point being positive.
[0189] Clause 24: A computer-readable device according to any one of clauses 19 to 23, wherein the first machine learning model includes a regressor and the second machine learning model includes a classifier, which is trained using a training data set including (i) the multipath component data, (ii) the true value location of the wireless device, and (iii) observability of the wireless device by a virtual access point.
[0190] Clause 25: A computer-readable device as described in any of clauses 19 to 24, wherein the output includes a heat map including indications of the one or more predicted wireless measurements for corresponding locations regarding a radio frequency (RF) environment associated with the wireless network.
[0191] Clause 26: A computer-readable device as described in any of clauses 19 to 25, wherein the location of the virtual access point is at a distance from a reflective surface to the virtual access point that is equal to the distance from the reflective surface to the location of the access point.
[0192] Clause 27: A computer-readable device according to any one of clauses 19 to 26, wherein the first machine learning model is further trained by generating training outputs, the training outputs comprising predicted multipath components; and the optimization comprises iterative minimization of the error between at least the predicted multipath components and the labeled multipath components.
[0193] Clause 28: A computerized device comprising: a component for obtaining a location of a user equipment (UE) of a wireless network; and a component for generating an output comprising one or more predicted wireless measurements using a first machine learning model given the location of the UE, wherein the first machine learning model is trained by utilizing at least the following items: a component for obtaining a training data set comprising (i) multipath component data and (ii) a true value location of a wireless device, wherein the multipath component data corresponds to a virtual access point associated with an access point of the wireless network, the virtual access point comprising a location determined based on the location of the access point; and a component for performing optimization with respect to the multipath component data and the true value location.
[0194] Clause 29: A computerized device according to clause 28, wherein: the multipath component data includes data corresponding to flight time, signal power, or a combination thereof; and the one or more predicted wireless measurements include flight time, signal power, or a combination thereof.
[0195] Clause 30: A computerized device according to any one of clauses 28 to 29, wherein the computerized device also includes a component for determining the observability of the virtual access point to the UE using a second machine learning model; the first machine learning model includes a regressor and the second machine learning model includes a classifier, the classifier is trained using a training data set including (i) the multipath component data, (ii) the true value location of the wireless device and (iii) the observability of the virtual access point to the wireless device; and the generation of the output including the one or more predicted wireless measurements is performed in response to the observability of the virtual access point being positive.
Claims
1. A method for predicting wireless channel data associated with a user equipment (UE) of a wireless network, the method comprising: Obtaining the location of the UE; as well as generating an output comprising one or more predicted wireless measurements using a first machine learning model given the location of the UE, the first machine learning model being trained by at least the following operations: obtaining a training data set including (i) multipath component data and (ii) a ground-truth location of a wireless device, the multipath component data corresponding to a virtual access point associated with an access point of the wireless network, the virtual access point including a location determined based on a location of the access point; as well as An optimization is performed with respect to the multipath component data and the true value positions.
2. The method according to claim 1, wherein: The multipath component data includes data corresponding to flight time, signal power, or a combination thereof; and The one or more predicted wireless measurements include time of flight, signal power, or a combination thereof.
3. The method of claim 1, wherein the method is performed by the UE, and the method further comprises receiving, by the UE, the first machine learning model from a network entity of the wireless network.
4. The method of claim 1, wherein the method is performed by a base station configured for data communication based on a Third Generation Partnership Project (3GPP), and the method further comprises transmitting, by the base station, the output including the one or more predicted wireless measurements to the UE.
5. The method of claim 1 , further comprising determining observability of the virtual access point to the UE using a second machine learning model; The generating of the output including the one or more predicted wireless measurements is performed in response to the observability of the virtual access point being positive.
6. The method of claim 5, wherein the first machine learning model comprises a regressor and the second machine learning model comprises a classifier, the classifier being trained using a training data set comprising (i) the multipath component data, (ii) the true value location of the wireless device, and (iii) observability of the wireless device by a virtual access point.
7. The method of claim 1, wherein the output comprises a heat map including indications of the one or more predicted wireless measurements for corresponding locations of a radio frequency (RF) environment associated with the wireless network.
8. The method of claim 1, wherein the location of the virtual access point is at a distance from a reflective surface to the virtual access point that is equal to a distance from the reflective surface to the location of the access point.
9. The method according to claim 1, wherein: The first machine learning model is further trained by generating training outputs, the training outputs comprising predicted multipath components; and The optimization comprises iterative minimization of an error between at least the predicted multipath component and the labeled multipath component.
10. The method according to claim 1, further comprising: Based on the predicted one or more wireless measurements, the position of the virtual access point is updated, or at least one of the one or more wireless measurements is provided to a server for use in positioning the UE.
11. A network node in a wireless network, the network node comprising: at least one wireless communication interface; Memory; as well as one or more processors, the one or more processors being communicatively coupled to the at least one wireless communication interface and the memory, and the one or more processors being configured to: obtaining a location of a user equipment (UE); and using a first machine learning model to generate an output comprising one or more predicted wireless measurements given the location of the UE; The first machine learning model is trained by at least the following operations: obtaining a training data set including (i) multipath component data and (ii) a ground-truth location of a wireless device, the multipath component data corresponding to a virtual access point associated with an access point of the wireless network, the virtual access point including a location determined based on a location of the access point; as well as An optimization is performed with respect to the multipath component data and the true value positions.
12. The network node according to claim 11, wherein: The multipath component data includes data corresponding to flight time, signal power, or a combination thereof; and The one or more predicted wireless measurements include time of flight, signal power, or a combination thereof.
13. The network node according to claim 11, wherein: The network node is configured for data communication based on the Third Generation Partnership Project (3GPP); and The one or more processors are further configured to transmit, by the network node, the output comprising the one or more predicted wireless measurements to the UE via the at least one wireless communication interface.
14. The network node according to claim 11, wherein: The one or more processors are further configured to determine observability of the virtual access point to the UE using a second machine learning model; and The generating of the output comprising the one or more predicted wireless measurements is performed in response to the observability of the virtual access point being positive.
15. The network node of claim 14, wherein the first machine learning model comprises a regressor and the second machine learning model comprises a classifier, the classifier being trained using a training data set comprising (i) the multipath component data, (ii) the true value location of the wireless device, and (iii) observability of the wireless device by a virtual access point.
16. The network node of claim 11, wherein the output comprises a heat map including indications of the one or more predicted wireless measurements for corresponding locations of a radio frequency (RF) environment associated with the wireless network.
17. The network node of claim 11, wherein the location of the virtual access point is at a distance from a reflective surface to the virtual access point that is equal to a distance from the reflective surface to the location of the access point.
18. The network node according to claim 11, wherein: The first machine learning model is further trained by generating training outputs, the training outputs comprising predicted multipath components; and The optimization comprises iterative minimization of an error between at least the predicted multipath component and the labeled multipath component.
19. A computer-readable device, the computer-readable device comprising a storage medium, the storage medium comprising a plurality of instructions, the plurality of instructions being configured to, when executed by one or more processors, cause a computerized device to: obtaining a location of a user equipment (UE) within a wireless network; and using a first machine learning model to generate an output comprising one or more predicted wireless measurements given the location of the UE; The first machine learning model is trained by at least the following operations: obtaining a training data set including (i) multipath component data corresponding to a virtual access point associated with an access point of the wireless network and (ii) a ground-truth location of a wireless device, the multipath component data corresponding to a virtual access point associated with an access point of the wireless network, the virtual access point including a location determined based on a location of the access point; and An optimization is performed with respect to the multipath component data and the true value positions.
20. The computer readable device of claim 19, wherein: The multipath component data includes data corresponding to flight time, signal power, or a combination thereof; and The one or more predicted wireless measurements include time of flight, signal power, or a combination thereof.
21. The computer readable device of claim 19, wherein: The computerized device comprises the UE; and The plurality of instructions are further configured to, when executed by the one or more processors, cause the UE to receive the first machine learning model from a network entity of the wireless network.
22. The computer readable device of claim 19, wherein: The computerized device includes a base station configured for data communications based on the Third Generation Partnership Project (3GPP); and The plurality of instructions are further configured to, when executed by the one or more processors, cause the base station to transmit the output including the one or more predicted wireless measurements to the UE.
23. The computer readable device of claim 19, wherein: The plurality of instructions are further configured to, when executed by the one or more processors, cause the computerized apparatus to determine observability of the virtual access point to the UE using a second machine learning model; and The generating of the output comprising the one or more predicted wireless measurements is performed in response to the observability of the virtual access point being positive.
24. A computer-readable device according to claim 23, wherein the first machine learning model comprises a regressor and the second machine learning model comprises a classifier, which is trained using a training data set including (i) the multipath component data, (ii) the true value location of the wireless device, and (iii) observability of the wireless device by a virtual access point.
25. The computer-readable device of claim 19, wherein the output comprises a heat map including indications of the one or more predicted wireless measurements for corresponding locations regarding a radio frequency (RF) environment associated with the wireless network.
26. The computer-readable device of claim 19, wherein the location of the virtual access point is at a distance from a reflective surface to the virtual access point that is equal to a distance from the reflective surface to the location of the access point.
27. The computer readable device of claim 19, wherein: The first machine learning model is further trained by generating training outputs, the training outputs comprising predicted multipath components; and The optimization comprises iterative minimization of an error between at least the predicted multipath component and the labeled multipath component.
28. A computerized device comprising: Means for obtaining a location of a user equipment (UE) of a wireless network; as well as Means for generating, given the location of the UE, using a first machine learning model to generate an output comprising one or more predicted wireless measurements, the first machine learning model trained by utilizing at least: means for obtaining a training data set comprising (i) multipath component data corresponding to a virtual access point associated with an access point of the wireless network and (ii) a ground-truth location of a wireless device, the multipath component data corresponding to a virtual access point associated with an access point of the wireless network, the virtual access point comprising a location determined based on a location of the access point; as well as Means for performing an optimization with respect to said multipath component data and said true value positions.
29. The computerized apparatus of claim 28, wherein: The multipath component data includes data corresponding to flight time, signal power, or a combination thereof; and The one or more predicted wireless measurements include time of flight, signal power, or a combination thereof.
30. The computerized apparatus of claim 28, wherein: The computerized apparatus further comprises means for determining observability of the virtual access point to the UE using a second machine learning model; The first machine learning model includes a regressor, and the second machine learning model includes a classifier, the classifier being trained using a training data set including (i) the multipath component data, (ii) the ground-truth location of the wireless device, and (iii) observability of the wireless device by a virtual access point; and The generating of the output comprising the one or more predicted wireless measurements is performed in response to the observability of the virtual access point being positive.
