Method and equipment for estimating position of user equipment (UE)

Abstract

The embodiment of the invention relates to a user equipment (UE) position estimation method and equipment. An exemplary method may include sending a first message to an LMF associated with one or more BSs of a wireless network; receiving a second message from the LMF in response to transmitting the first message, wherein the second message includes at least information for determining an AI model for UE position estimation; and receiving the AI model as determined from the LMF.

Description

Technical Field

[0001] Embodiments of the present application relate to wireless communication technology, and more particularly, to artificial intelligence (AI) applications in wireless communication, such as a method and apparatus for estimating a user equipment (UE) position. Background Art

[0002] AI, which includes at least machine learning (ML), is used to learn and perform certain tasks by training neural networks (NN) with large amounts of data, and has been successfully applied to the fields of computer vision (CV) and natural language processing (NLP). Deep learning (DL), a subordinate concept of ML, uses multiple layers of NN as an "AI model" to learn how to solve problems and / or optimize performance from large amounts of data.

[0003] By leveraging the benefits of AI, the performance of the Radio Access Network (RAN) network can be further optimized in at least the following use cases: energy saving, load balancing, traffic steering, and mobility optimization. Therefore, the Third Generation Partnership Project (3GPP) has been considering the introduction of AI into 3GPP since 2016, including several study items and work items in SA1, SA2, SA5, and RAN3. However, there are still a lot of issues that have not been discussed in 3GPP.

[0004] Taking UE location (or position) estimation as an example, in the current 3GPP discussion related to AI-based positioning, the Industrial Internet of Things (IIoT) indoor factory (InF) ​​scenario is a priority scenario for evaluating AI-based positioning, and the fingerprint recognition method is considered as an example of using an AI model to estimate the UE location based on the wireless environment (e.g., the reference signal received power (RSRP) of the associated reference signal). However, some issues should be addressed, such as how the UE obtains the AI ​​model used by fingerprint recognition, for example.

[0005] Therefore, the industry needs a technical solution for UE location estimation, which is based on an AI model from the core network (CN) side (e.g., location management function (LMF)) and will improve AI applications in further long term evolution (LTE) of RAN. Summary of the invention

[0006] One purpose of the embodiments of the present application is to provide a technical solution for wireless communications, especially a technical solution for UE position estimation based on an AI model.

[0007] Some embodiments of the present application provide a UE comprising: a processor, and a transceiver coupled to the processor, wherein the transceiver is configured to: send a first message to an LMF associated with one or more base stations (BSs) of a wireless network; receive a second message from the LMF in response to sending the first message, wherein the second message includes at least information for determining an AI model for UE position estimation; and receive the AI ​​model as determined from the LMF.

[0008] Some other embodiments of the present application provide an exemplary method that can be performed in a UE. The exemplary method includes: sending a first message to a LMF associated with one or more BSs of a wireless network; receiving a second message from the LMF in response to sending the first message, wherein the second message includes at least information for determining an AI model for UE position estimation; and receiving the AI ​​model as determined from the LMF.

[0009] In some embodiments of the present application, the first message is one of: a message indicating the UE's support for AI-based positioning estimation; a message requesting location service; a message requesting assistance data related to AI-based positioning estimation; or a message providing the positioning-related capabilities of the UE.

[0010] In some embodiments of the present application, the first message indicates at least one of the following: a target scenario; a preferred positioning method; or an AI capability of the UE.

[0011] In some embodiments of the present application, the information for determining the AI ​​model for UE position estimation includes information related to AI model transmission, and the information related to AI model transmission indicates at least one of the following: a uniform resource location (URL) owned by the LMF or a fully qualified domain name (FQDN) owned by the LMF, from which the AI ​​model will be downloaded; a type of learning method used by the AI ​​model; a type of the AI ​​model; parameters related to the AI ​​model; the entire payload of the AI ​​model; or an identification of the AI ​​model. According to some embodiments of the present application, receiving the AI ​​model from the LMF includes downloading the AI ​​model from the URL or the FQDN via a user plane protocol.

[0012] In some embodiments of the present application, the transceiver is configured to further receive information related to the application of the AI ​​model, and the information related to the application of the AI ​​model includes at least one of the following: information related to a reference signal, which is to be measured and the measurement result is to be used as an input to the AI ​​model; information related to the validity of the AI ​​model regarding the reference signal; information related to the validity of the AI ​​model regarding the area; or information related to the validity of the AI ​​model regarding the validity period.

[0013] According to some embodiments of the present application, in the case of configuring the reference signal to be measured, the information related to the application of the AI ​​model includes information for determining whether the AI ​​model is valid, wherein the information for determining whether the AI ​​model is valid includes at least one of the following: at least one indicator, wherein each indicator is associated with a reference signal and indicates whether the reference signal needs to be detected; or at least one threshold, each threshold being a threshold that the measured RSRP of one or more associated reference signals needs to be higher than or equal to.

[0014] According to some embodiments of the present application, information related to the AI ​​model application is received before the AI ​​model is downloaded from the LMF or after the AI ​​model is downloaded from the LMF. For example, in the case where information related to the AI ​​model application is received before the AI ​​model is downloaded from the LMF, the information related to the AI ​​model application is also included in the second message. In the case where information related to the AI ​​model application is received after the AI ​​model is downloaded from the LMF, the transceiver is configured to transmit a message after downloading the AI ​​model to request information related to the AI ​​model application.

[0015] In some embodiments of the present application, in response to at least one problem detected during the AI ​​model transfer procedure, the transceiver is configured to report at least one of the following reasons to the LMF: AI model download error; AI model compliance error; or AI model validity error.

[0016] In some embodiments of the present application, the transceiver is further configured to transmit at least one of the following to the LMF: information about common capabilities related to AI-based positioning; or information about capabilities related to one or more specific AI-based positioning methods. Exemplary information about common capabilities related to AI-based positioning includes at least one of the following: an indicator implying the maximum complexity or maximum size or maximum computing power of the AI ​​model supported by the UE; the maximum floating point operations per second (FLOPS) supported by the UE; the inference latency achieved by the UE when using a reference AI model; or the maximum number of AI models that the UE can store. Exemplary information about capabilities related to one or more specific AI-based positioning methods includes at least one of the following: supported frequency bands; supported frequencies; supported bandwidths; the maximum number of frequency layers supported by the UE; or the maximum number of reference signals processed by the UE in a time slot.

[0017] Some embodiments of the present application provide a LMF comprising: a processor; and a transceiver coupled to the processor, wherein the transceiver is configured to: receive a first message from a UE; transmit a second message to the UE in response to receiving the first message, wherein the second message includes at least information for determining an AI model for UE position estimation; and transmit the AI ​​model as determined to the UE.

[0018] In some embodiments of the present application, the transceiver is configured to further transmit information related to the AI ​​model application to the UE, and the information related to the AI ​​model application includes at least one of the following: information related to a reference signal, which is to be measured and the measurement result is to be used as an input to the AI ​​model; information related to the validity of the AI ​​model regarding the reference signal; information related to the validity of the AI ​​model regarding the area; or information related to the validity of the AI ​​model regarding the validity period.

[0019] According to some embodiments of the present application, the information related to the AI ​​model application is transmitted before the UE downloads the AI ​​model from the LMF or after the UE downloads the AI ​​model from the LMF. In the case where the information related to the AI ​​model application is transmitted before the UE downloads the AI ​​model from the LMF, the information related to the AI ​​model application is also included in the second message. In the case where the information related to the AI ​​model application is transmitted after the UE downloads the AI ​​model from the LMF, the transceiver is configured to: receive a message requesting the information related to the AI ​​model application after downloading the AI ​​model; and transmit the information related to the AI ​​model application in response to receiving the message.

[0020] In some embodiments of the present application, the transceiver is configured to receive at least one of the following causes from the UE, each cause indicating a problem detected during the AI ​​model transfer procedure: an AI model download error; an AI model compliance error; or an AI model validity error.

[0021] In some embodiments of the present application, the transceiver is further configured to transmit a message to request information about the positioning-related capabilities of the UE, and receive the information about the positioning-related capabilities of the UE, wherein the information about the positioning-related capabilities of the UE includes at least one of the following: information about common capabilities related to AI-based positioning; or information about capabilities related to one or more specific AI-based positioning methods.

[0022] In some embodiments of the present application, the payload of the AI ​​model is transmitted to the UE via a user plane protocol.

[0023] In summary, the embodiments of the present application propose a technical solution for UE position estimation, wherein an AI model for UE position estimation can be downloaded from the CN side via a user plane protocol, so that AI-based positioning can be implemented, which will improve the accuracy of UE position estimation and facilitate the implementation of AI-based RAN. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to describe the manner in which the advantages and features of the present application can be obtained, the description of the present application is presented by reference to specific embodiments of the present application illustrated in the accompanying drawings. These drawings depict only exemplary embodiments of the present application and are therefore not intended to limit the scope of the present application.

[0025] Figure 1 is a schematic diagram illustrating an exemplary wireless communication system according to some embodiments of the present application.

[0026] Figure 2 An exemplary architecture of a 5G system with location-related functions according to some embodiments of the present application is described.

[0027] Figure 3 is a flowchart illustrating an exemplary procedure of a method for UE position estimation according to some embodiments of the present application.

[0028] Figure 4 An exemplary procedure of a method for UE position estimation according to some embodiments of the present application is described.

[0029] Figure 5 Another exemplary procedure of a method for UE position estimation according to some other embodiments of the present application is described.

[0030] Figure 6 A block diagram illustrating a UE position estimation device according to some embodiments of the present application.

[0031] Figure 7 A block diagram illustrating a UE position estimation device according to some other embodiments of the present application. DETAILED DESCRIPTION

[0032] The detailed description of the accompanying drawings is intended as a description of the currently preferred embodiments of the present application, and is not intended to represent the only form in which the present application can be practiced. It should be understood that the same or equivalent functions can be implemented by different embodiments that are intended to be included in the spirit and scope of the present application.

[0033] Reference will now be made in detail to some embodiments of the present application, examples of which are illustrated in the accompanying drawings. To facilitate understanding, embodiments are provided under specific network architectures and new service scenarios (e.g., 3GPP 5G, 3GPP LTE, etc.). It should be considered that, with the development of network architectures and new service scenarios, all embodiments in the present application are also applicable to similar technical problems. In addition, the terms cited in the present application may change, which should not affect the principles of the present application.

[0034] Figure 1 A schematic diagram illustrating an exemplary wireless communication system 100 in accordance with some embodiments of the present application.

[0035] As in Figure 1 , the wireless communication system 100 includes at least one BS 101 and at least one UE 102. Specifically, for illustrative purposes, the wireless communication system 100 includes one BS 101 and two UEs 102 (e.g., a first UE 102a and a second UE 102b). Figure 1 A specific number of BSs and UEs are illustrated in FIG. 1 , but it is contemplated that in some other embodiments of the present application, the wireless communication system 100 may include more or fewer BSs and UEs.

[0036] The wireless communication system 100 is compatible with any type of network capable of sending and receiving wireless communication signals. For example, the wireless communication system 100 is compatible with wireless communication networks, cellular telephone networks, networks based on time division multiple access (TDMA), networks based on code division multiple access (CDMA), networks based on orthogonal frequency division multiple access (OFDMA), LTE networks, networks based on 3GPP, 3GPP 5G networks, satellite communication networks, high altitude platform networks, and / or other communication networks.

[0037] BS101 may communicate with a CN node (not shown) via an interface, such as a mobility management entity (MME) or a serving gateway (S-GW), a mobility management function (AMF), a user plane function (UPF) or a LMF, etc. BS may also be referred to as an access point, an access terminal, a base station, a macro cell, a Node B, an enhanced Node B (eNB), a gNB, a home Node B, a relay node or a device, or described using other terms used in the art. In 5G NR, BS may also refer to a RAN node or a network device. Each BS may serve several UEs within a service area (e.g., a cell or a cell sector) via a wireless communication link. Neighbor BSs may communicate with each other as needed, for example, during a handover procedure of a UE.

[0038] UE 102 (e.g., first UE 102a and second UE 102b) should be understood as any type of terminal device, which may include a computing device, such as a desktop computer, a laptop computer, a personal digital assistant (PDA), a tablet computer, a smart TV (e.g., a TV connected to the Internet), a set-top box, a game console, a security system (including a security camera), a vehicle onboard computer, a network device (e.g., a router, a switch and a modem), or the like. According to an embodiment of the present application, UE may include a portable wireless communication device, a smart phone, a cellular phone, a flip phone, a device with a subscriber identification module, a personal computer, a selective call receiver, or any other device capable of sending and receiving communication signals on a wireless network. In some embodiments, UE includes a wearable device, such as a smart watch, a fitness band, an optical head-mounted display, or the like. In addition, UE may be referred to as a subscriber unit, a mobile device, a mobile station, a user, a terminal, a mobile terminal, a wireless terminal, a fixed terminal, a subscriber station, a user terminal or a device, or described using other terms used in the art.

[0039] Traditionally, UE position estimation can be done in the UE side or in the LMF side, depending on the wireless communication deployment and scenario.

[0040] For example, to support positioning of target UEs and delivery of location assistance data to UEs with NG-RAN access in 5G systems (5GS), such as Figure 2 The location-dependent functions are distributed as shown in the architecture in TS 38.305. Figure 5 .1-1 and as described in more detail in TS 23.501 [2] and TS 23.273

[35] . The overall sequence of events for the UE, NG-RAN and LMF applicable to any location service is as follows: Figure 2 It is shown below.

[0041] First, a location service for the target UE will be requested from the LMF. For example, in step 201a, some 5GC location service (LCS) entity (e.g., Gateway Mobile Location Center (GMLC)) requests some location services (e.g., positioning) for the target UE from the AMF (e.g., Serving AMF). Or, in step 201b, the AMF for the target UE determines the need for some location services (e.g., to locate the UE for an emergency call). Or, in step 201b, the target UE requests some location services (e.g., positioning or delivery of auxiliary data) from the AMF at the non-access stratum (NAS) level.

[0042] It should be noted that when the AMF receives a location service request while the UE is in the Connection Management (CM) Idle state, the AMF performs a network triggered service request as defined in TS23.502 and TS23.273 to establish a signaling connection with the UE and assign a specific serving gNB or ng-eNB. Figure 2 that is, any signaling that may be required to put the UE into connected mode before step 201a is not shown. However, while positioning is still in progress, the signaling connection may be released later (e.g., by the NG-RAN node due to signaling and data inactivity).

[0043] In step 202, the AMF transmits a location service request to the LMF.

[0044] In step 203a, the LMF initiates location procedures with the serving and possibly neighbouring ng-eNB or gNB in ​​the NG-RAN, for example to obtain positioning measurements or assistance data.

[0045] In addition to or instead of step 203a, the LMF initiates a location procedure with the UE in step 203b, for example to obtain a position estimate or positioning measurements, or to transmit position assistance data to the UE.

[0046] In step 204, the LMF provides a location service response to the AMF and includes any required results, such as a success or failure indication, and, if requested and obtained, a location estimate for the UE.

[0047] In step 205a, if step 201a is performed, the AMF returns a location service response to the 5GC entity in step 201a and includes any required results, such as the UE's location estimate.

[0048] In step 205b, if step 201b occurs, the AMF uses the location service response received in step 204 to assist the service that triggered this in step 201b (for example, a location estimate associated with an emergency call to the GMLC may be provided).

[0049] In step 205c, if step 201c is performed, the AMF returns a location service response to the UE and includes any required results, such as the UE's location estimate.

[0050] When AI was introduced into 3GPP, AI-based UE position estimation was possible in the UE side in some scenarios. For example, in an IIoT scenario such as INF, the UE may be a robot moving around. If the UE can use an AI model to perform position estimation (e.g., by fingerprinting based on channel observations as input to the AI ​​model), it will be very helpful to improve positioning accuracy. However, in the legacy UE-based 5G positioning, the UE estimates its own position using auxiliary data received from the LMF via LPP messages, which are designed to transmit control plane messages of up to 1MB in size. Since the size of the AI ​​model may be as high as 100MB, which is much higher than 1MB, it is necessary to solve the problem of how the UE can obtain the AI ​​model (or positioning AI model) for UE position estimation.

[0051] Taking at least the above problems into consideration, embodiments of the present application provide a technical solution for UE position estimation based on an AI model transmitted from the CN side. For example, some embodiments of the present application provide a method and apparatus for UE position estimation, wherein an AI model transmitted (or transmitted) from a CN (e.g., from an LMF) via a user plane protocol is used for UE position estimation. The AI ​​model transmission procedure may be triggered due to a requested message or an unsolicited message, the requested message may be based on a request from the UE, such as an LPP RequestAssistanceData message, and the unsolicited message may be based on LPPProvideAssistanceData. The AI ​​model (e.g., an AI model for fingerprint recognition) may be trained by the LMF or other wireless devices using data collected from the UE, such as a set of ground truth locations and the measured RSRP at each ground truth location. In this article, it is assumed that the AI ​​model has been trained and stored in the LMF.

[0052] Figure 3 1 is a flowchart illustrating an exemplary procedure of a method for estimating a UE position according to some embodiments of the present application. Although the method is implemented at the system level by a remote device (e.g., Figure 1However, those skilled in the art may understand that the method implemented in the UE and the method implemented in the LMF may be implemented separately and combined by other devices with similar functions.

[0053] refer to Figure 3 , will request the AI ​​model related to the UE position estimation from the LMF for the UE, which can be implemented in various ways. For example, in step 301, the UE will, for example, request the AI ​​model related to the UE position estimation from the LMF for the UE, which can be implemented in various ways. Figure 2 The AMF of the AMF described in sends a first message to the LMF, where the LMF may be associated with one or more BSs (e.g., a serving BS of the UE). The first message is of various kinds and may indicate at least one of the following: a target scenario, a preferred positioning method (e.g., fingerprint recognition), or the AI ​​capability of the UE. For example, the first message may be a message indicating the UE's support for AI-based positioning estimation. In another instance, the first message is a message for a location service request, such as a NAS message for a location service request. In yet another instance, the first message is a message for requesting assistance data related to AI-based positioning estimation, such as an LPP message of RequestAssistanceData. In yet another instance, the first message is a message providing the positioning-related capabilities of the UE. In the case where the UE transmits the first message to the LMF via the AMF, the AMF may also select the LMF based on the first message.

[0054] Regarding the target area, it may be a geographic area as specified in TS23.273. For example, if desired, the target area is expressed in one of the following formats: a) a shape defined in TS23.032; b) a local coordinate system of the LCS client, such as using the Open Mobile Alliance (OMA) Mobile Location Protocol (MLP) protocol; c) an E.164 country code of the geographic area of ​​the LCS client, such as using the OMA MLP protocol; d) a public land mobile network (PLMN) identity of the LCS client, such as using the OMA MLP protocol; and e) a geopolitical name of the region of the LCS client (e.g., London), such as using the OMA MLP protocol.

[0055] Regarding the AI ​​capabilities of the UE, it may include positioning-related capabilities of the UE, such as common capabilities related to AI-based positioning and capabilities related to one or more specific AI-based positioning methods (e.g., fingerprinting). Exemplary information about common capabilities related to AI-based positioning may include at least one of the following: a) an indicator, such as low, medium, high, which implies the maximum complexity or maximum size or maximum computing power of the AI ​​model supported by the UE; b) the maximum FLOPS supported by the UE; c) the inference latency achieved by the UE when using a reference AI model (or a default AI model); or d) the maximum number of AI models that the UE can store. Information about capabilities related to one or more specific AI-based positioning methods may include processing capabilities related to reference signals, such as at least one of the following: frequency bands supported by the UE; frequencies supported by the UE; bandwidths supported by the UE; the maximum number of frequency layers supported by the UE; or the maximum number of reference signals processed by the UE in a time slot.

[0056] The UE may actively transmit the first message. For example, in some scenarios, sufficient measurements have been collected and stored in the UE. If the UE has determined that the amount of collected positioning measurements can be used as input into the AI ​​model, the UE may decide to request the AI ​​model for UE position estimation. The UE may transmit a general indicator to the LMF, such as transmitting a location service request message to the LMF or a message indicating the UE's support for AI-based positioning estimation. In some other scenarios, the LMF may decide to transmit the AI ​​model to the UE for UE position estimation. In such scenarios, the first message is a location service request message from the UE or a message indicating the UE's support for AI-based positioning estimation.

[0057] In some other embodiments of the present application, the first message is transmitted in response to receiving another message from the LMF or other device. For example, similar to the above, the UE may transmit a general indicator to the LMF, such as transmitting a location service request message to the LMF or a message indicating the UE's support for AI-based positioning estimation. The LMF may trigger a capability transmission procedure to further understand the UE's ability to support AI-based positioning, rather than directly deciding to transmit an AI model to the UE for UE position estimation. In some other scenarios, the LMF may trigger the capability transmission procedure for other reasons, and a message or the like indicating the UE's support for AI-based positioning estimation from the UE is not necessary. In the capability transmission procedure triggered by the LMF, the LMF may request the UE to provide the UE's AI capabilities by sending an LPP message of RequestCapabilities to the UE. In response to the request, the UE will transmit information about its AI capabilities to the LMF, such as an LPP message of ProvideCapabilities. In this capability transmission procedure triggered by the LMF, the first message may be an LPP message of ProvideCapabilities in response to the LPP message of RequestCapabilities.

[0058] In step 302, the LMF will receive a first message from the UE, for example, via a NAS message or an LPP message. In step 304, the LMF will provide a UE position estimation response to the UE in response to the first message, for example, transmitting a second message to the UE in response to the first information. In some scenarios, the LMF may determine whether to transmit an AI model for UE position estimation to the UE based on information of the UE (for example, based on the UE's ability to support AI-based positioning provided in the first message or other messages). In the case where the LMF determines that the AI ​​model for UE position estimation can be transmitted to the UE, the LMF will transmit a second message to the UE, the second message containing at least information for determining the AI ​​model for UE position estimation. In some other scenarios, the LMF may transmit a second message directly to the UE in response to reception of the first message.

[0059] In the case where the LMF determines that the UE is capable of AI-based UE position estimation, the second message will include at least information of the AI ​​model used to determine the UE position estimation. The LMF will transmit the determined AI model to the UE in step 306. Therefore, the UE will receive the second message from the LMF in step 305 and receive the AI ​​model from the LMF in step 307. Transmitting the AI ​​model from the LMF to the UE means transmitting the AI ​​model from the perspective of the LMF side and receiving the AI ​​model from the perspective of the UE side. Exemplary AI model transmission includes transmission of the payload of the AI ​​model, downloading the AI ​​model from a URL or FQDN owned by the LMF via a user plane protocol, and / or transmission of basic assistance data, such as information related to the validity of the AI ​​model, which may involve different parameters, such as about areas or time periods, etc. Exemplary AI model transmission may also include transmission of other information related to the AI ​​model in addition to the above.

[0060] In some other embodiments of the present application, the LMF may transmit a second message to the UE for reasons other than receiving the first message from the UE. Figure 2 In step 201a described in FIG. 2 , some 5GC LGS entities may transmit a first message to request some location services (eg, positioning) for the target UE from the LMF, such as location services via a serving AMF. Figure 2 In step 201b described in the figure, the serving AMF of the target UE may determine the need for some location services (e.g., for emergency call positioning UE), and then transmit a first message, such as a location service request, to the LMF. After receiving the first message, the LMF will transmit a second message including at least information for determining the AI ​​model for UE location estimation.

[0061] The information used to determine the AI ​​model used for UE position estimation may include information related to the transmission of the AI ​​model. Exemplary information related to the transmission of the AI ​​model may indicate at least one of the following: a) a URL owned by the LMF or a FQDN owned by the LMF, from which the AI ​​model will be downloaded, the URL or FQDN may be encoded as an OTCET string; b) the type of learning method used by the AI ​​model, such as unsupervised or supervised learning, clustering, classification, dimensionality reduction, regression, etc.; c) the type of AI model, such as a deep neural network, a support vector machine (SVM), a k-nearest neighbor (KNN) classifier, or the like; d) parameters related to the AI ​​model (rather than the entire payload of the AI ​​model), such as neural network weights, number of layers, support vectors (for SVM), and coefficients of linear or logistic regression, etc.; e) the entire payload of the AI ​​model; or f) the ID of the AI ​​model.

[0062] In some embodiments of the present application, the LMF may further transmit information related to the AI ​​model application to the UE. Some or all of the information related to the AI ​​model application to the UE may be collected by the LMF, for example, via the NR Positioning Protocol A (NRPPa) procedure or by other means from the relevant BS (e.g., gNB). The information related to the AI ​​model application may be transmitted (or received from the perspective of the UE side) before the AI ​​model is downloaded from the LMF or transmitted (or received from the perspective of the UE side) after the AI ​​model is downloaded from the LMF. For example, in the case where the information related to the AI ​​model application is received by the UE before the AI ​​model is downloaded from the LMF, the information related to the AI ​​model application may also be included in the second message.

[0063] Exemplary information related to the application of the AI ​​model may include information related to the reference signal to be measured, and the measurement results will be used as input to the AI ​​model. An exemplary reference signal may be a positioning reference signal (PRS), and the information related to the reference signaling may be a frequency layer, a transmit-receive point (TRP) identifier (ID), a PRS resource set ID, a PRS resource ID, and a related cell ID in terms of a physical cell identifier (PCI) or a NR cell global identifier (NCGI), or an absolute radio frequency channel number (ARFCN), etc. Another exemplary reference signal may be a synchronization signal (SS) and a physical channel (PBCH) block (SSB), and the information related to the reference signaling may be an SSB index and a related cell ID. Yet another exemplary reference signal may be a channel state information reference signal (CSI-RS), and the information related to the validity of the AI ​​model regarding the reference signal may be a CSI-RS index and a related cell ID.

[0064] In some other embodiments of the present application, the information related to the application of the AI ​​model may include information related to the validity of the AI ​​model regarding the reference signal, which may indicate that one or more reference signals that must be detected and / or the related RSRP or other measurement results must be above a threshold. For example, in the case of configuring a reference signal to be measured, the information related to the application of the AI ​​model may include information related to the validity of the AI ​​model regarding the reference signal, for example, information to determine whether the AI ​​model is valid. The information to determine whether the AI ​​model is valid may include at least one of the following: a) at least one indicator, each indicator being associated with a reference signal and indicating whether the reference signal needs to be detected; or b) at least one threshold, each threshold being a threshold that the measured RSRP of one or more associated reference signals needs to be higher than or equal to. For example, each reference signal is provided with a threshold, or multiple reference signals are provided with the same threshold.

[0065] The following shows an example of information related to the application of an AI model, where information related to a reference signal to be measured and the measurement results will be used as input to the AI ​​model, and includes information related to the effectiveness of the AI ​​model with respect to the reference signal.

[0066]

[0067]

[0068] According to the exemplary information related to the application of the AI ​​model, PRS, SSB and CSI-RS are configured, which will be measured and the measurement results will be used as input to the AI ​​model. In addition, an indicator associated with the PRS and the corresponding threshold is provided, which indicates that the PRS needs to be detected; and the RSRP of the detected RSRP should be higher than or equal to the threshold. Otherwise, the AI ​​model will be invalid for UE position estimation.

[0069] In other words, in the case of an AI model such as fingerprinting, the UE may use as input the measured RSRP of, say, 100 reference signals (not only PRS, but also SSB and CSI-RS), depending on the AI ​​model. However, among those input signals, some will be necessary, so that the UE location or position estimated by the AI ​​model is considered accurate or valid only when they can be detected or when they are good enough (e.g., the UE is close to the TRP).

[0070] In some other embodiments of the present application, the information related to the application of the AI ​​model may include information related to the validity of the AI ​​model for an area (e.g., an area defined in a cell list). For example, the information related to the validity of the AI ​​model for an area may indicate that the UE can use the AI ​​model only when the UE is connected to at least one of the provided cells.

[0071] In still other embodiments of the present application, the information related to the AI ​​model application may include information related to the validity of the AI ​​model regarding the validity period, for example, the period during which the measured reference signal can be applied to the AI ​​model. The information related to the validity of the AI ​​model regarding the validity period may be expressed in different ways. For example, the information related to the validity of the AI ​​model regarding the validity period may indicate the validity duration in units of "ms", "s", "minutes", or the like. In another example, the information related to the validity of the AI ​​model regarding the validity period may indicate an absolute time when the provided assistance data for the AI ​​model is considered to be expired or not applicable to the AI ​​model, such as Universal Coordinated Time (UTC).

[0072] Given that Figure 4 and 5Some more specific embodiments are described, in which Figure 4 In the embodiment, information related to the AI ​​application is received before the UE downloads the AI ​​model for UE position estimation from the LMF, and information related to the AI ​​application is received after the UE downloads the AI ​​model for UE position estimation from the LMF.

[0073] First, about Figure 4 , which illustrates an exemplary procedure of a method for UE position estimation according to some embodiments of the present application.

[0074] As in Figure 4 As shown in FIG. 4 , in step 401, the UE may transmit a first message to the LMF, as shown in FIG. Figure 3 The first message may be a NAS message, such as a location service request via AMF to LMF in step 401a, or an LPP message, such as a RequestAssistanceData message in step 401b. In some embodiments of the present application, the UE may transmit these two messages to the LMF.

[0075] In step 402, the LMF may transmit a second message to the UE in response to the first message. For example, the second message may be an LPP message, such as a ProvideAssistanceData message, which includes an information element (IE). The IE may be a new IE compared to the old one, and may contain only information related to the transmission of the AI ​​model, or both information related to the transmission of the AI ​​model and information related to the application of the AI ​​model. In the case where the second message does not include information related to the application of the AI ​​model, the LMF may transmit information related to the application of the AI ​​model to the UE in another message separate from the second message before the UE starts downloading the AI ​​model from the LMF.

[0076] After receiving information related to AI model transmission (e.g., the ID of the AI ​​model, and the URL or FQDN from which the AI ​​model can be downloaded) and information related to the AI ​​model application, the UE will start downloading the AI ​​model in step 403 based on the information related to AI model transmission via the user plane protocol, for example, from the provided URL or FQDN.

[0077] After downloading the AI ​​model (download completed), the UE will apply the relevant assistance data to the AI ​​model in step 404 and start UE position estimation.

[0078] about Figure 5 , which illustrates an exemplary procedure of a method for UE position estimation according to some other embodiments of the present application.

[0079] As in Figure 5 As shown in FIG. 5 , in step 501, the UE may transmit a first message to the LMF, as in Figure 3 The first message may be a NAS message, such as a location service request via AMF to LMF in step 501a, or an LPP message, such as a RequestAssistanceData message in step 501b. In some embodiments of the present application, the UE may transmit these two messages to the LMF.

[0080] In step 502, the LMF may transmit a second message to the UE in response to the first message. For example, the second message may be an LPP message, such as a ProvideAssistanceData message, which only includes information related to the AI ​​model transmission, such as the ID of the AI ​​model, and the URL or FQDN where the AI ​​model can be downloaded.

[0081] After receiving the information related to the AI ​​model transmission, the UE will start downloading the AI ​​model based on the information related to the AI ​​model transmission via a user plane protocol in step 503, for example from the provided URL or FQDN.

[0082] After downloading the AI ​​model, the UE will transmit another message to the LMF in step 504 to request information related to the AI ​​application, such as assistance data related to the AI ​​model. For example, the UE may transmit another LLP message, such as another RequestAssistanceData message, which includes the AI ​​model ID or other information that can identify the downloaded AI model. After receiving the request, the LMF will transmit information related to the AI ​​application of the AI ​​model to the UE in step 505. In another example, the UE may transmit another LPP message specifically for requesting AI positioning information or responding to AI positioning information, such as RequestAIMLAssistanceData, to request information related to the AI ​​application or respond to a second message, for example, which may contain an acknowledgment (ACK) or non-acknowledgment (NACK) indicating that the AI ​​model has been successfully downloaded or not successfully downloaded in addition to the AI ​​model ID. In the event that the AI ​​model is not successfully downloaded, an indicator may also be transmitted to the network (e.g., LMF) to indicate whether the AI ​​model has been partially downloaded (e.g., which may be indicated via a flag or a ratio of the downloaded AI payload to the total size of the AI ​​payload). In the case where the AI ​​model has been downloaded, the LMF will transmit information related to the AI ​​application to the UE in step 505. In some other embodiments of the present application, the LMF may determine whether the UE has downloaded the AI ​​model (whether the download is complete) regardless of whether there is a relevant message, such as a request or ACK from the UE (step 504 is not required). In the case where the LMF determines that the UE has downloaded the AI ​​model, the LMF will transmit information related to the AI ​​application to the UE. In some scenarios, information related to the AI ​​application may be collected before receiving a request or similar message. In other scenarios, the LMF will only start collecting information related to the AI ​​application from the relevant gNB after receiving a request or similar message.

[0083] After receiving the information related to the AI ​​application, the UE will apply the relevant assistance data to the AI ​​model in step 506 and start UE position estimation.

[0084] During the AI ​​model transmission procedure, some problems may occur. In response to at least one problem detected in the UE during the AI ​​model transmission procedure, the UE may report the corresponding cause to the LMF, for example, by sending an LPP abort message or an error message to the LMF. For example, the cause may be an AI model download error, which indicates, for example, that the AI ​​model download failed. In another example, the cause may be an AI model compliance error, which indicates, for example, that a given AI model cannot be adopted by the UE. In yet another example, the cause may be an AI model validity error, which indicates, for example, that the criteria for indicating model validity are not met. Exemplary scenarios in which the model validity criteria are not met may be that the UE moves out of the validity area, or that a basic reference signal (a reference signal configured to be measured) is not detected, or that the AI / ML model validity timer expires. This timer may be configured by the network (e.g., LMF) to indicate the validity duration of the AI ​​model.

[0085] In addition to the method for estimating the UE position, some embodiments of the present application also provide a UE position estimation device. For example, Figure 6 A block diagram illustrating a UE position estimation apparatus 600 according to some embodiments of the present application.

[0086] As in Figure 6 As shown, the apparatus 600 may include at least one non-transitory computer-readable medium 601, at least one receiving circuit system 602, at least one transmitting circuit system 604, and at least one processor 606 coupled to the non-transitory computer-readable medium 601, the receiving circuit system 602, and the transmitting circuit system 604. The at least one processor 606 may be a CPU, a DSP, a microprocessor, etc. The apparatus 600 may be a RAN node configured to perform the method described above or the like.

[0087] Although in this figure, elements such as at least one processor 606, transmit circuit system 604, and receive circuit system 602 are described in the singular, the plural may be considered unless explicitly limited to the singular. In some embodiments of the present application, receive circuit system 602 and transmit circuit system 604 may be combined into a single device, such as a transceiver. In certain embodiments of the present application, apparatus 600 may further include an input device, a memory, and / or other components.

[0088] In some embodiments of the present application, the non-transitory computer-readable medium 601 may store thereon computer-executable instructions to enable the processor to implement the method for the remote device (e.g., UE) as described above. For example, when the computer-executable instructions are executed, the processor 606 interacts with the receiving circuit system 602 and the transmitting circuit system 604 to perform the steps for the remote device as described above, for example, Figure 3 Displayed in.

[0089] In some embodiments of the present application, the non-transitory computer-readable medium 601 may store thereon computer-executable instructions to enable the processor to implement the method for the CN node (e.g., LMF) as described above. For example, when the computer-executable instructions are executed, the processor 606 interacts with the receiving circuit system 602 and the transmitting circuit system 604 to perform the steps for the CN node as described above, such as Figure 3 Displayed in.

[0090] Figure 7 is a block diagram of an apparatus for UE position estimation 700 according to some other embodiments of the present application.

[0091] refer to Figure 7 , the apparatus 700 (e.g., UE or LMF) may include at least one processor 702 and at least one transceiver 704 coupled to the at least one processor 702. The transceiver 704 may include at least one separate receive circuit system 706 and transmit circuit system 708, or at least one integrated receive circuit system 706 and transmit circuit system 708. The at least one processor 702 may be a CPU, a DSP, a microprocessor, etc.

[0092] According to some embodiments of the present application, when the device 700 is a remote device (e.g., UE), the UE includes a processor and a transceiver coupled to the processor, wherein the transceiver is configured to: send a first message to a LMF associated with one or more BSs (e.g., one or more gNBs) of a wireless network; receive a second message from the LMF in response to sending the first message, wherein the second message includes at least information for determining an AI model for UE position estimation; and receive the AI ​​model as determined from the LMF.

[0093] According to some other embodiments of the present application, when the device 700 is a CN node (e.g., LMF), the LMF includes a processor and a transceiver coupled to the processor, wherein the transceiver is configured to: receive a first message from an LMF associated with one or more BSs (e.g., one or more gNBs) of a wireless network; transmit a second message from the LMF in response to sending the first message, wherein the second message includes at least information for determining an AI model for UE position estimation; and transmit the AI ​​model as determined from the LMF.

[0094] The method according to the embodiment of the present application case can also be implemented on a programmed processor. However, the controller, flow chart and module can also be implemented on a general or special-purpose computer, a programmed microprocessor or microcontroller and peripheral integrated circuit elements, an integrated circuit, a hardware electronic or logic circuit (such as a discrete element circuit), a programmable logic device or the like. In general, any device capable of implementing the flow chart shown in the figure can be used to implement the processor function of the present application case. For example, an embodiment of the present application case provides a device including a processor and a memory. Computer programmable instructions for implementing the method are stored in a memory, and the processor is configured to execute the computer programmable instructions to implement the method. The method may be a method as stated above or other methods according to an embodiment of the present application case.

[0095] Alternative embodiments preferably implement the method according to the embodiments of the present application in a non-transitory computer-readable storage medium storing computer programmable instructions. The instructions are preferably executed by a computer executable component that is preferably integrated with a network security system. The non-transitory computer-readable storage medium can be stored on any suitable computer-readable medium, such as RAM, ROM, flash memory, EEPROM, optical storage device (CD or DVD), hard drive, floppy disk drive, or any suitable device. The computer executable component is preferably a processor, but the instructions can be executed alternatively or additionally by any suitable dedicated hardware device. For example, an embodiment of the present application provides a non-transitory computer-readable storage medium in which computer programmable instructions are stored. The computer programmable instructions are configured to implement the method as stated above or other methods according to the embodiments of the present application.

[0096] In addition, in the present disclosure, the term "includes / including" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or equipment comprising a series of elements not only comprises the elements, but also may comprise other elements that are not explicitly listed or inherent to the process, method, article or equipment. In the absence of more constraints, elements beginning with "one" or the like do not exclude the presence of additional equivalent elements in the process, method, article or equipment comprising the elements. In addition, the term "another" is defined as at least the second or more. The term "having" and the like as used herein are defined as "comprising".

Claims

1. A user equipment UE, comprising: processor; and a transceiver coupled to the processor, wherein the transceiver is configured to: sending a first message to a location management function LMF associated with one or more base stations BS of the wireless network; receiving a second message from the LMF in response to sending the first message, wherein the second message includes at least information for determining an artificial intelligence (AI) model for UE position estimation; and The AI ​​model as determined is received from the LMF.

2. The UE according to claim 1, wherein: The first message indicates at least one of the following: Target scenario; Preferred positioning method; or The AI ​​capability of the UE.

3. The UE according to claim 1, wherein: The information of the AI ​​model for determining the UE position estimate includes information related to the AI ​​model transmission, and the information related to the AI ​​model transmission indicates at least one of the following: A uniform resource location URL owned by the LMF or a fully qualified domain name FQDN owned by the LMF from which the AI ​​model will be downloaded; The type of learning method used by the AI ​​model; The type of the AI ​​model; Parameters related to the AI ​​model; The entire payload of the AI ​​model; or The identification of the AI ​​model.

4. The UE according to claim 1, wherein: The transceiver is configured to further receive information related to the AI ​​model application, and the information related to the AI ​​model application includes at least one of the following: Information related to a reference signal that will be measured and the measurement results will be used as input to the AI ​​model; Information related to the effectiveness of the AI ​​model with respect to the reference signal; Information related to the effectiveness of AI models with respect to the region; or Information related to the effectiveness of the AI ​​model regarding the effectiveness period.

5. The UE according to claim 4, wherein: In case of configuring the reference signal to be measured, the information related to the AI ​​model application includes information for determining whether the AI ​​model is valid, wherein the information for determining whether the AI ​​model is valid includes at least one of the following: at least one indicator, wherein each indicator is associated with a reference signal and indicates whether detection of the reference signal is required; or At least one threshold, each threshold being a threshold that the measured reference signal received power RSRP of one or more associated reference signals needs to be higher than or equal to.

6. The UE according to claim 4, wherein: The information related to the AI ​​model application is received before the AI ​​model is downloaded from the LMF or after the AI ​​model is downloaded from the LMF.

7. The UE according to claim 6, wherein: In the case where the information related to the AI ​​model application is received before downloading the AI ​​model from the LMF, the information related to the AI ​​model application is also included in the second message.

8. The UE according to claim 6, wherein: In the event that the information related to the AI ​​model application is received after downloading the AI ​​model from the LMF, the transceiver is configured to transmit a message to request the information related to the AI ​​model application after downloading the AI ​​model.

9. The UE according to claim 1, wherein: In response to at least one problem detected during the AI ​​model transfer procedure, the transceiver is configured to report at least one of the following reasons to the LMF: AI model download error; AI model compliance errors; or AI model validity errors.

10. The UE according to claim 1, wherein: The transceiver is further configured to transmit to the LMF at least one of: Information about common capabilities related to AI-based targeting; or Information about capabilities associated with one or more specific AI-based positioning methods.

11. The UE according to claim 10, wherein: The information about common capabilities related to AI-based positioning includes at least one of the following: an indicator implying a maximum complexity or a maximum size or a maximum computational power of the AI ​​model supported by the UE; The maximum floating point operations per second (FLOPS) supported by the UE; the inference latency achieved by the UE when using a reference AI model; or The maximum number of AI models that the UE can store.

12. The UE according to claim 10, wherein: The information regarding capabilities related to one or more specific AI-based positioning methods includes at least one of the following: frequency bands supported by the UE; frequencies supported by the UE; a bandwidth supported by the UE; a maximum number of frequency layers supported by the UE; or The maximum number of reference signals processed by the UE in a timeslot.

13. The UE according to claim 3, wherein: Receiving the AI ​​model from the LMF includes downloading the AI ​​model from the URL or the FQDN via a user plane protocol.

14. A location management function LMF, comprising: processor; and a transceiver coupled to the processor, wherein the transceiver is configured to: receiving a first message from a user equipment UE; In response to receiving the first message, transmitting a second message to the UE, wherein the second message includes at least information for determining an artificial intelligence (AI) model for UE position estimation; as well as The AI ​​model as determined is transmitted to the UE.

15. A method performed in a user equipment UE, comprising: sending a first message to a location management function LMF associated with one or more base stations BS of the wireless network; receiving a second message from the LMF in response to sending the first message, wherein the second message includes at least information for determining an artificial intelligence (AI) model for UE position estimation; and The AI ​​model as determined is received from the LMF.

Images