Artificial intelligence positioning method and wireless communication device

Abstract

The invention provides an artificial intelligence (AI) positioning method, which is executed in user equipment and comprises the following steps of: receiving AI positioning auxiliary data and / or model input reporting configuration from core network equipment, and reporting AI positioning information according to the AI positioning auxiliary data and / or model input reporting configuration. The invention also provides an artificial intelligence AI positioning method, which is executed in core network equipment and comprises the following steps of: sending AI positioning auxiliary data and / or model input report configuration to user equipment (UE); and receiving AI positioning information generated by the UE according to the AI positioning auxiliary data and / or model input report configuration.

Description

Artificial intelligence positioning method and wireless communication device Technical Field

[0001] The present invention relates to the field of communication systems, and more particularly to an artificial intelligence positioning method and wireless communication equipment. Background Art

[0002] With the rapid development of location-based services (LBS), precise positioning services can facilitate users' various activities and daily life needs. LBS includes applications such as tracking, navigation, information access, and mobile advertising. These applications and services must be based on high-accuracy positioning technology, and therefore require real-time and high-precision positioning technology to support them.

[0003] In related technologies, a terminal can perform positioning measurements on wireless signals emitted by multiple base stations or transmission reception points (TRPs), obtain positioning measurement data, and report the positioning measurement data to the location management function (LMF) entity on the network side. After the LMF entity determines the location of the terminal based on the positioning measurement data, it returns the determined location to the terminal. Based on the scenario's requirements for positioning accuracy, 3GPP studied positioning enhancement based on artificial intelligence (AI) in the R18 stage. The study showed that AI positioning requires the introduction of new measurement quantities to support the operation of the AI ​​model. At present, the measurement-related signaling form, including measurement configuration and measurement quantity reporting, has not been clarified. The signaling overhead of reporting the input of the sample-based model is large, and the amount of reported data needs to be compressed. Since positioning is a continuous process, especially when the positioning cycle is short, the measurement will occupy more time domain resources, and the measurement time needs to be shortened to reduce resource occupancy. Due to the limited generalization capability of the model, auxiliary information needs to be exchanged between the UE / gNB and LMF entities to ensure the consistency of model training and model inference. Therefore, it is necessary to clarify the auxiliary information and design the corresponding signaling process.

[0004] Technical Solution

[0005] An object of the present invention is to provide an artificial intelligence positioning method and a wireless communication device to solve the above technical problems.

[0006] A first aspect of the present invention provides an artificial intelligence (AI) positioning method, which is executed in a user device and includes: receiving AI positioning assistance data and / or model input reporting configuration from a core network device; and reporting AI positioning information according to the AI ​​positioning assistance data and / or model input reporting configuration.

[0007] A second aspect of the present invention provides an artificial intelligence (AI) positioning method, which is executed in a core network device and includes: sending AI positioning assistance data and / or model input reporting configuration to a user equipment (UE); and receiving AI positioning information generated by the UE based on the AI ​​positioning assistance data and / or model input reporting configuration.

[0008] A third aspect of the present invention provides an artificial intelligence (AI) positioning method, which is executed in a base station and includes: receiving AI positioning assistance data and / or model input reporting configuration from a core network device; and reporting AI positioning information according to the AI ​​positioning assistance data and / or model input reporting configuration.

[0009] The fourth aspect of the present invention provides an artificial intelligence (AI) positioning method, which is executed in a core network device and includes: sending AI positioning assistance data and / or model input reporting configuration to a base station; and receiving AI positioning information generated by the base station based on the AI ​​positioning assistance data and / or model input reporting configuration.

[0010] The method disclosed in the present invention can be implemented in a chip. The chip may include a processor configured to call and run a computer program stored in a memory, so that a device equipped with the chip executes the method disclosed in the present application.

[0011] The method disclosed in the present invention can be programmed as computer-executable instructions stored in a non-transitory computer-readable medium. When the non-transitory computer-readable medium is loaded into a computer, it instructs the processor of the computer to execute the method disclosed in the present invention.

[0012] The non-transitory computer readable medium may include at least one of the following readable media: a hard disk, a CD-ROM, an optical storage device, a magnetic storage device, a read-only memory, a programmable read-only memory, an erasable programmable read-only memory, an EPROM, an electrically erasable programmable read-only memory, and a flash memory.

[0013] The methods disclosed in the present invention may be programmed as a computer program product, which causes a computer to execute the methods disclosed in the present application.

[0014] The methods disclosed in the present invention may be programmed as computer programs, which cause a computer to execute the methods disclosed in the present application.

[0015] The method disclosed in the present invention can be implemented by a wireless communication device. The wireless communication device includes a processor and a memory, the memory is used to store a computer program, and the processor is used to call and run the computer program stored in the memory. Beneficial effects

[0016] Compared to the prior art, an embodiment of the present invention provides an artificial intelligence positioning method, which is executed in a user equipment (UE), and includes receiving AI positioning assistance data and / or a model input reporting configuration from a core network device, and reporting AI positioning information based on the AI ​​positioning assistance data and / or the model input reporting configuration. Another embodiment of the present application provides an artificial intelligence positioning method, which is executed in a core network device, and includes sending AI positioning assistance data and / or a model input reporting configuration to a user equipment (UE), and receiving AI positioning information generated by the UE based on the AI ​​positioning assistance data and / or the model input reporting configuration. Another embodiment of the present application provides an artificial intelligence positioning method, which is executed in a base station, and includes receiving AI positioning assistance data and / or a model input reporting configuration from a core network device, and reporting AI positioning information based on the AI ​​positioning assistance data and / or the model input reporting configuration. Another embodiment of the present application provides an artificial intelligence positioning method, which is executed in a core network device, and includes sending AI positioning assistance data and / or a model input reporting configuration to a base station, and receiving AI positioning information generated by the base station based on the AI ​​positioning assistance data and / or the model input reporting configuration. This application can improve the efficiency of AI positioning by introducing measurement configuration and measurement quantity reporting for AI positioning. At the same time, this application can also reduce reporting signaling overhead by compressing the amount of reported data. In addition, this application can reduce resource usage by shortening the measurement time. In addition, the UE / base station and core network equipment of this application exchange auxiliary information to ensure the consistency of model training and model inference. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0018] FIG1 is a schematic diagram illustrating the architecture of a wireless communication system according to the present invention.

[0019] FIG2 is a block diagram showing a wireless communication system according to the present invention including a UE, a base station and core network equipment.

[0020] FIG3 is a schematic diagram of a UE, a base station, and a LMF entity used for AI positioning in a wireless communication system according to an embodiment of the present invention.

[0021] FIG4A is a flow chart of a method for a UE to perform AI positioning according to an embodiment of the present invention.

[0022] FIG4B is a flowchart of a method for a UE to perform AI positioning according to an embodiment of the present invention.

[0023] FIG5 is a schematic diagram of a base station and an LMF entity used for AI positioning in a wireless communication system according to an embodiment of the present invention.

[0024] FIG6 is a flow chart of a method for a base station to perform AI positioning.

[0025] FIG. 7 is a schematic diagram illustrating sampling point selection performed by a UE / base station and an LMF entity.

[0026] FIG8 is a schematic diagram illustrating measurement value screening performed by a UE / base station and a LMF entity.

[0027] FIG9 is a schematic diagram illustrating the reconfiguration of reference signals by the LMF before and after the beam transmission sequence is changed.

[0028] Figure 10 is a schematic diagram of the UE and LMF entity operations on the UE side model training and inference consistency method.

[0029] Figure 11 is a schematic diagram of the UE and LMF entities operating on the LMF side model training and inference consistency method.

[0030] Figure 12 is a schematic diagram of the base station and LMF entity operating the LMF side model training and inference consistency method. Modes for Carrying Out the Invention

[0031] The embodiments of the present application describe in detail the technical matters, structural features, implementation objectives and effects with reference to the accompanying drawings. Specifically, the terms in the embodiments of the present application are only used for the purpose of describing specific embodiments, rather than limiting the disclosure.

[0032] In the present invention, "A or B" may mean "only A", "only B" or "both A and B".

[0033] In other words, in the present invention, "A or B" can be interpreted as "A and / or B". For example, in the present invention, "A, B or C" can mean "only A", "only B", "only C" or "any combination of A, B, and C".

[0034] A slash ( / ) or a comma used in the present invention may mean "and / or". For example, "A / B" may mean "A and / or B". Thus, "A / B" may mean "only A", "only B", or "both A and B". For example, "A, B, C" may mean "A, B, or C".

[0035] In the present invention, “at least one of A and B” may mean “only A”, “only B”, or “both A and B”. In addition, in the present invention, the expression “at least one of A or B” or “at least one of A and / or B” may be interpreted as “at least one of A and B”.

[0036] In addition, in the present invention, "at least one of A, B, and C" may mean "only A", "only B", "only C", or "any combination of A, B, and C". In addition, "at least one of A, B, or C" or "at least one of A, B, and / or C" may mean "at least one of A, B, and C".

[0037] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the described features. Throughout the description of this application, "plurality" means two or more, unless otherwise specifically defined.

[0038] Those skilled in the art will recognize and appreciate that the details of the described examples are merely illustrative of some embodiments and that the teachings set forth herein are applicable to various alternative arrangements.

[0039] The technical solution of the present invention can be applied to various wireless communication systems, such as: Long Term Evolution (LTE) system, LTE Frequency Division Duplex (FDD) system, LTE Time Division Duplex (TDD) system, 5G communication system or future wireless communication system. 5G communication system or 5G network can also be called New Radio (NR) system or NR network.

[0040] For example, a wireless communication system 100 to which the present invention is applied is shown in FIG1 . The wireless communication system 100 may include a core network 130, a base station 200, and a user equipment 10. The base station 200 may be a device that communicates with the user equipment (UE) 10. The base station 200 may provide communication coverage for a specific geographical area and may communicate with the user equipment 10 located within the coverage area.

[0041] Core network 130 may be an IP mobile communications network operated by a mobile communications operator. For example, core network 130 may be a core network used by a mobile communications operator that operates and manages wireless communication system 100, or may be a core network used by a virtual mobile communications operator such as a Mobile Virtual Network Operator (MVNO). Core network 130 may be connected to base station 200 and serve as a relay device for transmitting user data. User device 10 transmits and receives user data via core network 130. It should be noted that user data communication is not limited to IP communication and may also involve non-IP communication.

[0042] Optionally, the base station 200 can be an evolved base station (eNB) in an LTE system, or the base station can be a mobile switching center, a relay station, an access point, a vehicle-mounted device, a wearable device, a hub, a switch, a bridge, a router, a network side device in a 5G network, or a base station in a future communication system, etc.

[0043] Optionally, the UE 10 may be stationary or mobile. The user equipment 10 includes, but is not limited to, a connection via a wired line, such as via a Public Switched Telephone Network (PSTN), a Digital Subscriber Line (DSL), a digital cable, a direct cable connection; and / or another data connection / network; and / or via a wireless interface, such as for a cellular network, a Wireless Local Area Network (WLAN), a digital television network such as a DVB-H network, a satellite network, an AM-FM broadcast transmitter; and / or a device of another user equipment configured to receive / send communication signals; and / or an Internet of Things (IoT) device. A user equipment configured to communicate via a wireless interface may be referred to as a "wireless communication terminal," "wireless terminal," or "mobile terminal." Examples of mobile terminals include, but are not limited to, satellite or cellular telephones; Personal Communications System (PCS) terminals that can combine cellular radio telephones with data processing, fax, and data communications capabilities; may include radiotelephones, pagers, personal digital assistants (PDAs), wireless modems, wireless communication devices, handheld devices, laptop computers, tablet devices, cameras, gaming devices, netbooks, smartbooks, ultrabooks, medical devices or apparatuses, wearable devices (smart watches, smart clothing, smart glasses, smart wristbands), entertainment devices (music or video devices), vehicle-mounted components or sensors, smart meters / sensors, industrial manufacturing equipment, Global Positioning System (GPS) devices, or any other appropriate device configured to communicate via wireless or wired media. The access terminal can be a cellular phone, a cordless phone, a Session Initiation Protocol (SIP) phone, a Wireless Local Loop (WLL) station, a personal digital assistant, a handheld device with wireless communication capabilities, a computing device or other processing device connected to a wireless modem, an in-vehicle device, a wearable device, a user device in a 5G network, or a user device in a future evolved PLMN, etc.

[0044] Alternatively, two or more UEs (e.g., UE 10) may communicate directly using one or more sidelink channels (e.g., without using a base station as an intermediary for communicating with each other). For example, the UE 10 may communicate using point-to-point (P2P) communication, device-to-device (D2D) communication, vehicle-to-everything (V2X) protocols (which may include vehicle-to-vehicle (V2V) protocols, vehicle-to-infrastructure (V2I) protocols, or similar protocols), mesh networks, or similar networks, or a combination thereof. In this case, the UE 10 may perform scheduling operations, resource selection operations, and other operations described elsewhere herein as being performed by a base station.

[0045] In the embodiment of the present invention, the base station 200 can perform uplink (UL) transmission and downlink (DL) transmission with the user equipment 10 .

[0046] As shown in Figure 2, the communication system includes user equipment (UE) 10, a base station 200, and a core network device 300. The connections between the devices and device components are shown as lines and arrows in the figure. The UE 10 may include a processor 11, a memory 12, and a transceiver 13. The base station 200 may include a processor 201, a memory 202, and a transceiver 203. The core network device 300 may include a processor 301, a memory 302, and a transceiver 303. Each processor 11, 201, 301 may implement the functions, processes, and / or methods provided in the embodiments after execution. The wireless interface protocol layer may be implemented in the processor 11, 201, 301. Each memory 12, 202, 302 may store various programs and information to cooperate with the operation of the connected processor. Each transceiver 13, 203, 303 is coupled to the processor and is used to send and / or receive radio signals or wired signals. The base station 200 can be an eNB, a gNB, an access point (AP), a transmit-receive point (TRP), or one of other types of wireless nodes, and can configure wireless resources for the UE 10.

[0047] Each processor 11, 201, 301 may include an application-specific integrated circuit (ASIC), other chipsets, logic circuits, and / or data processing devices. Each memory 12, 202, 302 may include a read-only memory (ROM), a random access memory (RAM), flash memory, a memory card, a storage medium, and / or other storage devices. Each transceiver 13, 203, 303 may include a baseband circuit and a radio frequency (RF) circuit to process radio frequency signals. When the embodiment is implemented in software, the technology described herein may be implemented by executing the functional modules, processes, functions, entities, etc. described herein. The module may be stored in memory and executed by the processor. The memory may be implemented inside or outside the processor, and various devices known in the art may be coupled to the processor.

[0048] In this embodiment, the core network device 300 may be a node in the LTE core network or the 5G core network 130, including a user plane function (UPF), a session management function (SMF), a mobility management function (AMF), a unified data management (UDM), a policy control function (PCF), a control plane (CP) / user plane (UP) separation (CUPS), an identity authentication server (AUSF), a network slice selection function (NSSF), and a network exposure function (NEF). The core network device 300 serves as a location management function (LMF) entity (hereinafter referred to as LMF).

[0049] The present application provides methods for downlink AI positioning and uplink AI positioning. Wherein downlink AI positioning involves signaling interaction between the LMF entity 300 and the UE 10, while uplink AI positioning involves signaling interaction between the LMF entity 300 and the base station 200. On the one hand, the present application ensures the normal operation of the model by introducing measurement configuration related to AI positioning and reporting forms of model input (sample-based and path-based). On the other hand, the three methods of differential reporting, sampling point screening, and measurement value screening are used to compress the amount of data input to the model to reduce the signaling overhead of reporting. On the other hand, the measurement time window is configured based on the AI ​​model capability and auxiliary information to reduce the measurement time. On the other hand, the signaling process for the interaction of auxiliary information between the UE and LMF and the signaling process for the interaction of auxiliary information between the base station and LMF are designed to ensure the consistency of model training and model reasoning.

[0050] Please refer to Figures 1, 2, 3, and 4A-4B. Figure 3 is a schematic diagram of a UE, a base station, and a LMF entity used for AI positioning in a wireless communication system according to an embodiment of the present invention. Figure 4A is a flow chart of a method for a UE to perform AI positioning according to an embodiment of the present invention, and Figure 4B is a flow chart of a method for a UE to perform AI positioning according to an embodiment of the present invention. As shown in Figure 3, the wireless communication system performing AI positioning may include the following steps:

[0051] Step S301: AI positioning assistance data interaction is requested by the UE or actively provided by the LMF entity.

[0052] Step S302: The base station 200 configures the MG / PPW for the UE 10 and reduces the configured measurement duration based on the model capability and the auxiliary information configuration.

[0053] Step S303: LMF requests AI positioning information from UE, and requests the required model input measurement values, reporting method, and auxiliary information by introducing AI positioning related measurement configuration.

[0054] Step S304: UE 10 performs PRS measurement and / or model inference according to the LMF configuration.

[0055] Step S305: After measurement and / or model inference, the UE 10 reports AI positioning information to the LMF, introducing model input, such as sample-based model input and path-based model input reporting.

[0056] It is worth noting that the above steps may be executed sequentially or non-sequentially, and furthermore, only some steps may be executed instead of executing all steps every time.

[0057] As shown in FIG4A , in one embodiment, the method for a UE to perform AI positioning includes the following steps:

[0058] Step S400: Receive AI positioning assistance data and / or model input reporting configuration from the core network device.

[0059] Step S406: reporting AI positioning information according to the AI ​​positioning assistance data and / or model input reporting configuration.

[0060] As shown in FIG4B , in another embodiment, the method for a UE to perform AI positioning includes the following steps:

[0061] Step S400: Receive AI positioning assistance data and / or model input reporting configuration from the core network device.

[0062] Step S402: Receive a measurement window duration configuration for AI positioning from a base station.

[0063] Step S406: reporting AI positioning information according to the AI ​​positioning assistance data and / or model input reporting configuration.

[0064] It is worth noting that the above steps may be performed sequentially or not, and may be performed only partially instead of all at once. Therefore, compared to FIG4A , step S402 shown in FIG4B is an optional step.

[0065] Please refer to Figures 1, 2, 5, and 6. Figure 5 is a schematic diagram of a base station and an LMF entity in a wireless communication system for AI positioning according to an embodiment of the present invention. Figure 6 is a flow chart of a method for a base station to perform AI positioning. As shown in Figure 5, the wireless communication system performing uplink AI positioning may include the following steps:

[0066] Step S501: The LMF requests positioning-related information from the base station and configures the required SRS configuration parameters.

[0067] Step S502: LMF requests measurement from the base station, and by introducing AI positioning-related measurement configuration, requests the required model input measurement values, reporting methods, and auxiliary information.

[0068] Step S503: The base station performs SRS measurement and / or intermediate measurement value inference according to the LMF configuration.

[0069] Step S504: After measurement and / or model inference, the base station reports the AI ​​positioning information to the LMF.

[0070] It is worth noting that the above steps may be executed sequentially or non-sequentially, and furthermore, only some steps may be executed instead of executing all steps every time.

[0071] As shown in FIG6 , the method for performing AI positioning by a base station includes the following steps:

[0072] Step S600: Receive AI positioning assistance data and / or model input reporting configuration from the core network device.

[0073] Step S601: reporting AI positioning information according to the AI ​​positioning assistance data and / or model input reporting configuration.

[0074] In step 301, AI positioning assistance data exchange is requested by UE 10 or proactively provided by LMF entity 300. The AI ​​positioning assistance data includes at least one of the following: a positioning reference signal (PRS) resource subset for AI positioning, on-demand PRS pre-configuration, a measurement threshold, or a model validity area.

[0075] In step 302, UE 10 receives a measurement window duration configuration for AI positioning from base station 200, wherein the measurement window duration configuration for AI positioning indicates a symbol length, and the measurement window duration configuration includes two types: measurement gap (MG) and PRS processing window (PPW).

[0076] If the AI ​​model is deployed in the Location Management Function (LMF) entity 300, the LMF entity 300 configures the required measurement and reporting model input types for the UE 10 and / or base station 20 based on the model function for model training, model monitoring, and model inference. The model input types that the LMF entity 300 requests the UE 10 and / or base station 20 to report are sample-based model input and / or path-based model input.

[0077] When the LMF entity 300 requests a sampling point-based model input from the UE 10 and / or the base station 20 , it configures one or more of the following parameters: a data type parameter, a sampling point number parameter, or a sampling granularity parameter.

[0078] The data type parameter indicates the type of channel response sequence to be reported, including at least one of the following: Delay Profile (DP), Power Delay Profile (PDP), Channel Impulse Response (CIR), Power Spectrum (Power Spectrum), or Channel Frequency Response (CFR). DP represents the delay sequence corresponding to the sampling points with higher power in the channel response. PDP represents the delay and power sequence of the channel response. CIR represents the delay, power, and phase sequence of the channel response. PS represents the frequency and power sequence of the channel response. CFR represents the frequency, power, and phase sequence of the channel response.

[0079] The sampling point number parameter indicates the number of sampling points that need to be reported. When the LMF entity 300 configures the sampling point number parameter, it instructs the UE 10 and / or the base station 20 to select the sampling points with higher power for reporting.

[0080] The sampling granularity parameter indicates the granularity of filtering the sampling points to be reported. The sampling granularity parameter is an integer. When the LMF entity 300 configures the sampling granularity parameter and its value is p, if the data type parameter is a time domain sequence, i.e., DP, PDP, and CIR, the UE 10 and / or base station 20 selects the first 1 / p sampling points. If the data type is a frequency domain sequence, i.e., PS and CFR, the UE 10 and / or base station 20 uses the first sampling point as the starting point and selects sampling points every p-1 interval.

[0081] When the data type parameter is configured as DP, the number of sampling points parameter and / or the sampling granularity parameter must be configured. If both the number of sampling points and the sampling granularity parameter are configured, the UE 10 or base station 20 first filters sampling points based on the sampling granularity and then filters sampling points based on the number of sampling points. If neither the number of sampling points nor the sampling granularity parameter is configured, the UE 10 or base station 20 reports all sampling point information.

[0082] When LMF entity 300 requests path-based model input from UE 10, it configures one or more of the following parameters: data type, number of multipaths, delay type, or granularity factor. When LMF entity 300 requests path-based model input from base station 20, it configures one or more of the following parameters: data type, number of multipaths, or granularity factor.

[0083] The data type parameter indicates the reported multipath parameters, including at least one of the following: delay, delay power, or delay power phase. The multipath number parameter indicates the number of multipaths to be reported. The delay type parameter indicates the reporting format of the multipath delay, which can be configured as relative and / or absolute. The granularity factor parameter indicates the resolution of the reported multipath delay.

[0084] The relative form requires the delay of a certain multipath of a certain base station as a reference. For example, the delay of the first multipath of the reference TRP can be used as a reference. The other multipath delays of the TRP and the multipath delays of other TRPs are relative delays to the reference delay. The absolute form reports the absolute delay of each multipath.

[0085] In addition, the LMF entity 300 can also configure the training type to the UE 10 and / or the base station 20 to indicate the reporting content of the UE 10 and / or the base station 20, wherein the model training type configuration is used to indicate whether to report the positioning position and / or positioning measurement value. The model training type includes supervised learning and / or semi-supervised learning. When the LMF entity 300 configures the model training type to the UE 10, supervised learning indicates that the UE 10 needs to report one or more of the following information while reporting the model input: non-radio access technology (Radio Access Technology, RAT) positioning position (such as GPS coordinates), RAT positioning position (such as coordinates obtained by DL-TDOA technology) or RAT positioning measurement value (such as DL RSTD); unsupervised learning indicates that the UE can only report the model input. When the LMF entity 300 configures the training type to the base station 20, supervised learning indicates that the base station 20 needs to report at least the RAT positioning measurement value (such as UL RTOA) while reporting the model input. Unsupervised learning indicates that the base station 20 can only report the model input.

[0086] According to the above description, in downlink positioning, an example of the LMF entity 300 sending a measurement configuration to the UE 10 through LTE Positioning Protocol (LPP) signaling is as follows:

[0087] According to the above description, in uplink positioning, an example in which the LMF entity 300 sends a measurement configuration to the base station 20 through the New Radio Positioning Protocol (NR Positioning Protocol annex, NRPPa) signaling is as follows:

[0088] The UE 10 and / or the base station 20 reports the model input to the LMF entity 300 and / or a non-3GPP entity (e.g., an over-the-top (OTT) service platform) for model inference, model training, model monitoring, and other functions. The model input is obtained through the channel response of the reference signal and can be divided into a sample-based model input and a path-based model input. The sampling point-based model input is a sampling sequence of the channel response, such as DP, PDP, CIR, PS, and CFR, and its sampling time is an integer multiple of the sampling period. The path-based model input contains information about each multipath, such as delay, power, and phase, and its sampling time is not necessarily an integer multiple of the sampling period.

[0089] When the UE 10 and / or the base station 20 needs to report the model input based on the sampling points to the LMF entity 300, in order to save signaling overhead and assist the LMF entity 300 in completing the channel response reproduction, it is necessary to set the sampling rules of the UE 10 and / or the base station 20. That is, the time interval between sampling points is determined based on the bandwidth of the reference signal (e.g., positioning reference signal (PRS)) or the frequency interval between sampling points is determined based on the subcarrier spacing of the reference signal, and the number of complete sampling points is not less than the minimum power of 2 of the number of reference signal resource elements (REs).

[0090] When the LMF entity 300 has configured the number of sampling points, the UE 10 or the base station 20 may use a bitmap indication and / or an ordinal indication of the sampling points.

[0091] When the bitmap is indicated, the number of bits is the number of complete sampling points, and the filtered sampling points are set to 1, the sampling points not reported are set to 0, and the number of bits with a value of 1 is equal to the number of configured sampling points. At the same time, the parameters (sampling values) corresponding to each sampling point are reported in the order of the sampling points with bits set to 1.

[0092] The sampling point ordinal number indication directly indicates the ordinal number of the filtered and reported sampling points, and the parameters corresponding to each sampling point are reported in the order of the filtered sampling points. If the LMF entity 300 is not configured with the number of sampling points, the UE 10 and / or base station 20 only needs to report the parameters corresponding to each sampling point and does not need to indicate the sampling points.

[0093] The sampling point parameters that need to be reported depend on the data type parameter configured by the LMF entity 300 based on the sampling point model input. When the data type parameter is DP, the UE 10 and / or base station 20 does not need to report the parameters corresponding to each sampling point. When the data type parameter is PDP or PS, the UE 10 and / or base station 20 needs to report the power corresponding to each sampling point. When the data type parameter is CIR or CFR, the UE 10 and / or base station 20 needs to report the power and phase corresponding to each sampling point.

[0094] The parameters corresponding to the sampling points can be reported using bit string indications and / or mapping table indexes. The bit string indication indicates that a floating point number is represented by a bit string. When reporting the power corresponding to the sampling point, a set of bit strings, such as 32 bits, is used to represent the power. When reporting the power and phase corresponding to the sampling point, two sets of bit strings are used to simultaneously indicate the power-phase combination. One set of bit strings indicates the real part of the complex number corresponding to the power-phase combination, and the other set of bit strings indicates the imaginary part of the complex number corresponding to the power-phase combination, such as 32 bits representing the real part of the complex number and 32 bits representing the imaginary part of the complex number. The mapping table index indicates that the parameters are divided into multiple intervals, each interval corresponding to an integer. The corresponding parameter interval can be obtained by the reported integer. The power and phase corresponding to the sampling point are reported using the power mapping table and the phase mapping table, respectively.

[0095] In downlink positioning, the UE 10 reports the model input based on the sampling point to the LMF entity 300 through LPP signaling, and reports the measured PRS resource ID at the same time. The LMF entity 300 device 300 derives the delay resolution based on the PRS bandwidth or the frequency resolution through the subcarrier spacing to determine the delay or frequency corresponding to each sampling point. In addition, when the sampling point parameters are reported in the form of a mapping table index, the reporting granularity of the power and phase of the model input based on the sampling point needs to be small enough to retain the channel information and ensure the accuracy of the UE 10 position inferred by the model. Regarding power, the PRS received sampling point power (positioning reference signal received sample power, PRS-RSRSP) can be defined to represent the sampling point power of the channel response corresponding to the resource element carrying the PRS, and a predefined PRS-RSRSP reporting mapping table can be used to index the power. For example, the PRS-RSRSP reporting mapping table shown in Table 1 has a range of -156 to -31 dBm and a granularity of 0.5 dBm. The table can be used in at least one of the following ways:

[0096] 1. The LMF entity 300 indicates the table to be used, including at least one of the following: 3GPP TS 38.133 Table 10.1.24.3.1-1, 3GPP TS 38.133 Table 10.1.38.3.1-1, or a predefined PRS-RSRSP reporting mapping table (such as Table 1);

[0097] 2. At least one of the following tables is used by default: 3GPP TS 38.133 Table 10.1.24.3.1-1, 3GPP TS 38.133 Table 10.1.38.3.1-1, or a predefined PRS-RSRSP reporting mapping table (such as Table 1).

[0098] Table 1

[0099] According to an embodiment of the present application, an example of an LPP signaling message in a bitmap indication mode is as follows:

[0100] According to an embodiment of the present application, an example of an LPP signaling message in a sampling point ordinal indication mode is as follows:

[0101] According to one embodiment of the present application, in uplink positioning, the base station 200 reports the model input based on the sampling point to the LMF entity 300 through NRPPa signaling, and reports the positioning reference signal (sounding reference signal, SRS) resource ID and / or positioning SRS resource ID. The LMF entity 300 derives the delay resolution of the model input according to the SRS bandwidth or the frequency resolution according to the subcarrier spacing to indicate the delay or frequency corresponding to each sampling point. Similarly, when the sampling point parameters are reported in the form of a mapping table index, in order to ensure the accuracy of the AI ​​model inferring the position of UE 10, SRS-RSRSP is defined to represent the sampling point power of the channel response corresponding to the resource element carrying SRS, and a predefined SRS-RSRSP reporting mapping table is used to index the power, such as the SRS-RSRSP reporting mapping table shown in Table 2, the range of SRS-RSRSP is -156 to -31dBm, and the granularity is 0.5dBm. The table can be used in at least one of the following ways:

[0102] 1. The LMF entity 300 indicates the table to be used, which is at least one of the following: 3GPP TS 38.133 Table 13.3.1-1, 3GPP TS 38.133 Table 13.6.1-1, or a predefined SRS-RSRSP reporting mapping table (such as Table 2);

[0103] 2. At least one of the following tables is used by default: 3GPP TS 38.133 Table 13.3.1-1, 3GPP TS 38.133 Table 13.6.1-1, or a predefined SRS-RSRSP reporting mapping table (such as Table 2).

[0104] Table 2

[0105] Based on the above description, an example of NRPPa signaling message is as follows:

[0106] According to another embodiment of the present application, the path-based model input, namely the multipath parameters, includes the multipath delay, power, and phase. Based on these three parameters, three input combinations can be formed, namely, delay, delay power, and delay power phase, which correspond to DP, PDP, and CIR in the sampling point-based model input, respectively.

[0107] The reporting granularity of the power and phase inputs of the path-based model needs to be small enough to retain more channel information and ensure the accuracy of the UE position inferred by the model. For example, the granularity of power is 0.5dBm.

[0108] In addition, since the path-based model input loses some channel information compared to the sample-based model input, the UE 10 or the base station 20 can report at least one of the following parameters while reporting the multipath parameters: delay quality, power (RSRP) quality or phase quality to enhance the performance of the AI ​​model, where the RSRP quality is the error of the parameter estimation algorithm, indicating the estimate of RSRP uncertainty. This parameter is a non-negative real number in dBm.

[0109] During downlink positioning, UE 10 reports path-based model inputs to LMF entity 300 via LPP signaling. The delay type parameter is associated with multipath delay and includes absolute and / or relative forms. The relative form indicates the relative delays of the multipath delay of a base station, with the multipath delay of the base station as the reference delay, and the multipath delays of the other multipath delays of the base station and the multipath delays of multiple base stations other than the base station. The absolute form reports the absolute multipath delays of multiple base stations.

[0110] Specifically, the absolute delay is defined as the time of receiving the subframe containing PRS from a certain base station relative to the DL-RTOA reference time (Reference Time) of the base station, where the DL-RTOA reference time is defined as T0+t PRS , T0 represents the start time of the first system frame SFN#0 of the transmitting point, t PRS =(10n f +n sf )×10 -3 Indicates the starting time of the subframe where the PRS is located relative to the first system frame (SFN#0), where n f Indicates the frame number, n sf Indicates the subframe number. In order to save the reporting overhead of absolute delay, the delay index can be obtained through the predefined DL-RTOA reporting mapping table. The reporting range of DL-RTOA in the predefined table is -985024T c ~985024T c , the reported resolution is T=Tc ×2 k , T c is the time unit of NR, k is selected by the UE from the set {0, 1, 2, 3, 4, 5}, and the selected value is reported to the LMF entity 300. An example of the DL-RTOA reporting mapping table with k=0 is shown in Table 3.

[0111] Table 3

[0112] Relative delay is defined as the difference between the current delay and the reference delay. In order to save the reporting overhead of relative delay, the delay index can be performed through the predefined relative delay reporting mapping table for DL-RTOA. The reporting range of relative delay is -8175T c ~8175T c , the reported resolution is T=T c ×2 k , T c is the time unit of NR, k is selected by the UE in the set {0,1,2,3,4,5}, and the selected value is reported to the LMF entity 300. When indexing at this resolution, for example, when k=1, the UE reports that the Δpath is between -1 and 1 and uses the same index value. The LMF entity 300 cannot distinguish whether the multipath is before or after the reference multipath, resulting in distortion of the reported multipath delay information, thereby affecting the accuracy of AI positioning. Therefore, a method of locally improving the granularity of UE reporting delay is adopted, such as splitting -1≤Δpath<1 into two ranges of -1≤Δpath<0 and 0≤Δpath<1 for indexing. An example of a reporting mapping table for relative delay of DL-RTOA when k=1 is shown in Table 4.

[0113] Table 4

[0114] UE 10 can report path-based model inputs using at least one of the following methods: 1. Relative delay based on the Assistance Data Reference TRP (TRP); 2. Relative delay based on the minimum multipath delay; 3. Absolute delay for all multipaths within each TRP; 4. Absolute delay for the first multipath within each TRP. Note that the Transmit / Receive Point (TRP) mentioned below is equivalent to the base station in this case.

[0115] Option 1: Relative delay based on assistance data reference TRP: Taking the absolute delay of the first multipath of the assistance data reference TRP as a reference, the reporting delay of the first multipath is always 0, and the reporting delay of other multipaths of the TRP and the multipaths of other TRPs are relative delays with the first multipath. The delays of all multipaths are indexed by a predefined relative delay reporting mapping table for DL-RTOA (such as Table 4). In this way, when the UE reports the multipath delay, it needs to report at least one of the following information: multipath power, multipath phase, PRSID of each TRP, resource group ID of the reference TRP, resource ID under the resource ID group of the reference TRP, absolute delay of the first multipath of the reference TRP, physical cell identity (Primary Cell Identity, PCI) of each TRP, NR Global Cell Identity (NGCI) of each TRP or absolute radio frequency channel number (ARFCN) of each TRP, where the absolute delay of the first multipath of the reference TRP is indexed by Table 3, and the reporting of multipath power and / or multipath phase depends on the data type configured by the LMF entity 300.

[0116] Based on the above description, an example of an LPP signaling message is as follows: wherein the RSRP quality value (RSRPQualityValue) reported by UE 10 indicates an estimate of RSRP uncertainty in dBm, and the reported RSRP quality value resolution (RSRPQualityResolution) indicates the resolution, where mdot1 and m1 represent 0.1 dBm and 1 dBm, respectively.

[0117] Option 2: Relative delay based on minimum multipath delay: Taking the multipath corresponding to the minimum delay as a reference, the reporting delay of the multipath is always 0, and the reporting delays of other multipaths of the TRP corresponding to the multipath and the multipaths of other TRPs are relative delays to the minimum delay. Since the relative delay must be a positive value, the relative delay reporting mapping table for DL-RTOA can remove the negative part to reduce the signaling overhead of the reporting indication. The example of the reporting mapping table when k=1 is as follows. All multipath delays are indexed by a predefined relative delay reporting mapping table for DL-RTOA (such as Table 4 or Table 5).

[0118] Table 5

[0119] When UE 10 reports the multipath delay, it needs to report at least one of the following information: multipath power, multipath phase, minimum absolute delay, PRS ID of each TRP, PCI of each TRP, NCGI of each TRP or ARFCN of each TRP, where the minimum absolute delay is indexed by Table 3, and the reporting of multipath power and / or multipath phase depends on the data type configured by the LMF entity 300.

[0120] Based on the above description, an example of an LPP signaling message is as follows.

[0121] Option 3: Absolute delay for all multipaths in each TRP: TRP i The absolute delay of the corresponding multipath is referenced by its corresponding DL-RTOA reference time, TRP i The corresponding DL-RTOA reference time is defined as T i +t PRS,i , T i Indicates TRP i The starting time of the first SFN (SFN#0) can be obtained by the SFN offset from the reference TRP, t PRS,i =(10n f +n sf )×10 -3 Indicates TRP i The starting time of the subframe where the PRS is located relative to its SFN#0, where n f Indicates the frame number, n sf Indicates the subframe sequence number. All multipath delays are indexed by a predefined DL-RTOA reporting mapping table (e.g., Table 3). In this manner, when the UE reports the multipath delay, it needs to report at least one of the following information: multipath power, multipath phase, PRS ID of each TRP, PCI of each TRP, NCGI of each TRP, or ARFCN of each TRP. The reporting of multipath power and / or multipath phase depends on the data type configured by the LMF entity 300.

[0122] Based on the above description, an example of an LPP signaling message is as follows.

[0123] Option 4: The first multipath delay of each TRP is absolute: indexed by a predefined DL-RTOA reporting mapping table (such as Table 3). The remaining multipath delays of each TRP are relative to the first multipath, indexed by a predefined DL-RTOA relative delay reporting mapping table (such as Table 4 or Table 5). In this method, the UE needs to report at least one of the following information when reporting multipath delay: the PRS ID of each TRP, the PCI of each TRP, the NCGI of each TRP, or the ARFCN of each TRP.

[0124] Based on the above description, an example of LPP signaling message is as follows:

[0125] In uplink positioning, the base station 200 reports the path-based model input to the LMF entity 300 through NRPPa signaling. The reporting method is that the first multipath of each TRP uses the absolute delay, indexed by 3GPP TS 38.133 Table 13.1.1-1–Table 13.1.1-6, and the remaining multipath delays use the relative delay with the first multipath, indexed by 3GPP TS 38.133 Table 13.1.1A-1–Table 13.1.1A-6. The absolute delay (UL-RTOA) of each TRP corresponding to the first multipath is referenced to the UL-RTOA reference time, which is defined as T0+t SRS , T0 represents the starting time of the first SFN (SFN#0), t SRS =(10n f +n sf )×10 -3 Indicates the starting time of the SRS subframe relative to SFN#0, where n f Indicates the frame number, n sf Indicates the subframe number.

[0126] Thus, an example of NRPPa signaling message is as follows.

[0127] According to one embodiment of the present application, because the sampling point-based model input has a larger data volume than the path-based model input, in order to reduce the reporting overhead, the data volume of the model input can be compressed by at least one of the following methods: differential reporting, sampling point screening, or measurement value screening.

[0128] Option 1: Differential reporting

[0129] When the measurement time interval is short, the change in channel response is small. Therefore, when the model input is PDP, CIR, PS, or CFR, the model input at the next moment can be reported in differential form based on the previously reported model input.

[0130] The indication of the reported sampling point adopts the method of bitmap indication and / or sampling point ordinal indication. Taking into account that the phase is affected by noise and its change is irregular, the absolute phase of each sampling point is reported in the form of mapping table index. The power reporting of each sampling point adopts the method of differential reporting, that is, the differential component of the power with the same sampling point at the previous moment is reported. At this time, the mapping table of RSRSP differential reporting can be predefined, and the differential component is reported through the index table. Since the power change is very small in a short time, the number of optional differential components can be reduced to reduce signaling overhead. In the form of reporting partial sampling points, the sampling points filtered for reporting may change compared with the previous moment. If the currently reported sampling point was not reported at the previous moment, it is still reported in the form of absolute power. When the model input is reported for the first time, the reported sampling point is indicated by bitmap indication and / or sampling point ordinal indication, and the phase and power corresponding to the sampling point are reported in the form of mapping table index.

[0131] In downlink positioning, the predefined PRS-RSRSP differential reporting mapping table is shown in Table 6. The differential PRS-RSRSP range is -4 to 4dBm, with a granularity of 0.5dB. The table can be used in at least one of the following ways:

[0132] 1. The LMF entity 300 indicates the table to be used, which is at least one of the following: 3GPP TS 38.133 Table 10.1.24.3.2-2, 3GPP TS 38.133 Table 10.1.38.3.2-1, or a predefined PRS-RSRSP differential reporting mapping table (such as Table 6);

[0133] 2. At least one of the following tables is used by default: 3GPP TS 38.133 Table 10.1.24.3.2-2, 3GPP TS 38.133 Table 10.1.38.3.2-1, or a predefined PRS-RSRSP differential reporting mapping table (such as Table 6).

[0134] Table 6

[0135] Based on the above description, an example of an LPP signaling message is as follows.

[0136] In uplink positioning, a predefined SRS-RSRSP differential reporting mapping table may be used, as shown in Table 7, where the differential SRS-RSRSP range is -4 to 4 dBm, with a granularity of 0.5 dB.

[0137] Table 7

[0138] Option 2: Sampling point screening

[0139] Considering that only some time domain sampling points or frequency domain sampling points have large power fluctuations in a short period of time, it is possible to selectively report some sampling points with large power changes to reduce the signaling overhead of reporting.

[0140] Please refer to FIG. 7 , which is a schematic diagram illustrating sampling point screening performed by a UE / base station and an LMF entity.

[0141] In step S701, the LMF entity 300 requests measurement from the UE 10 via LPP signaling and / or requests measurement from the base station 200 via NRPPa signaling, and configures a sampling point power change threshold that indicates a condition for not reporting the power change of the sampling point.

[0142] An example of an LPP signaling message in downlink positioning is as follows, where the threshold (ThresholdValue) indicates the power change threshold of the sampling point in dBm, the threshold resolution (ThresholdResolution) indicates its resolution, and mdot1 and m1 represent 0.1 dBm and 1 dBm respectively.

[0143] An example of NRPPa signaling message in uplink positioning is as follows:

[0144] In step S702, the UE 10 reports the sampling point-based model input to the LMF entity 300 via LPP signaling, and / or the base station 200 reports the sampling point-based model input to the LMF entity 300 via NRPPa signaling. If this is the first time the model input is reported, the reported sampling point is indicated using a bitmap indication and / or a sampling point ordinal indication, and the phase and power corresponding to the sampling point are reported using a mapping table index. If this is not the first time the model input is reported, and the current sampling point is a newly filtered sampling point or the absolute value of the power difference between the current sampling point and the previous sampling point is greater than a threshold, the sampling point is indicated using a bitmap indication or a sampling point ordinal indication, and the corresponding parameters of the sampling point are reported. The reported parameters depend on the data type configured by the LMF entity 300; otherwise, the sampling point is not indicated and the corresponding parameters are not reported. When using a bitmap indication, the number of bits corresponds to the number of complete sampling points. Reported sampling points are set to 1, and unreported sampling points are set to 0. The sampling point ordinal number indication directly indicates the ordinal number of the reported sampling point. The power and / or phase of the newly filtered sampling point can be reported using a mapping table index. The phase of the sampling point whose absolute power error is greater than the threshold can be reported using a mapping table index. The power (RSRSP) of the sampling point whose absolute power error is greater than the threshold can be reported using at least one of the following methods:

[0145] 1. Indicated by a predefined PRS-RSRSP reporting mapping table and / or SRS-RSRSP reporting mapping table index;

[0146] 2. The model input compression reporting method is performed in a differential reporting manner through a predefined PRS-RSRSP differential reporting mapping table and / or an SRS-RSRSP differential reporting mapping table index indication.

[0147] 3. Based on method 2, the granularity of the differential reporting mapping table is determined according to the threshold configured by the LMF entity 300. The predefined PRS-RSRSP differential reporting mapping in this method is shown in Table 8, and the SRS-RSRSP differential reporting mapping is shown in Table 9, where ΔP represents the configured sampling point power change threshold.

[0148] Table 8

[0149] Table 9

[0150] Option 3: Measurement value filtering

[0151] If the measurement results of multiple resource sets are reported at the same reporting time, some model inputs can be filtered from the measurement results corresponding to different resource sets for reporting. Please refer to Figure 8, which shows a schematic diagram of measurement value screening between UE / base station and LMF entity.

[0152] In step S801, the LMF entity 300 requests measurement from the UE through LPP signaling and / or requests measurement from the base station 200 through NRPPa signaling, and configures the time domain / frequency domain channel correlation threshold and / or the minimum reporting quantity to indicate the conditions for model input screening reporting.

[0153] An example of the LPP signaling message in downlink positioning is as follows, where dot5, dot6, dot7, dot8, dot9, and 1 represent time domain / frequency domain channel correlation thresholds of 0.5, 0.6, 0.7, 0.8, 0.9, and 1, respectively.

[0154] An example of NRPPa signaling message in uplink positioning is as follows:

[0155] In step S802, the UE 10 reports the sampling point-based model inputs to the LMF entity 300 via LPP signaling, or the base station 200 reports the sampling point-based model inputs to the LMF entity 300 via NRPPa signaling. If the LMF entity 300 is configured only with time / frequency domain channel correlation thresholds, the UE 10 and / or base station 20 may calculate channel correlations and filter model inputs based on their own implementation. For example, using the first reported model input as a reference, the UE 10 and / or base station 20 may calculate the correlations of subsequent model inputs with the reference model input. If the calculated correlation is not less than the threshold, the model input is not reported. Instead, the model input is reported and used as a reference for calculating the correlations of subsequent model inputs. If the LMF entity 300 is configured only with a minimum reporting quantity, the UE 10 and / or base station 20 may filter the model inputs based on their own implementation for reporting, for example, filtering and reporting based on the measured PRS-RSRP and / or SRS-RSRP from highest to lowest. If the LMF entity 300 configures the time domain / frequency domain channel correlation threshold and the minimum reporting quantity at the same time, screening is performed based on the UE 10 or the base station 20 itself. For example, when the number of model inputs filtered by the correlation threshold does not meet the minimum reporting quantity, the remaining model inputs are screened from large to small according to the measured PRS-RSRP and / or SRS-RSRP.

[0156] Downlink positioning measurement enhancement

[0157] Please refer to Figure 3. To enhance the measurement result of downlink positioning, in step S302 of Figure 3, the base station 200 configures the MG / PPW for the UE 10 and reduces the configured measurement duration based on the model capability or auxiliary information configuration.

[0158] Option 1: Reduce measurement time without auxiliary information

[0159] Assuming that the model has strong generalization ability, that is, the UE measurement results corresponding to any beam pair can be used as the input of the model, the network side can configure a shorter measurement window for the UE based on the current PRS configuration to reserve more time-frequency resources for transmitting other signals, while also reducing the reporting overhead of the model input.

[0160] Considering that the number of symbols occupied by PRS can be 2, 4, 6, or 12, in the AI ​​positioning use case, the LMF entity 300 can configure the actual symbol-level measurement window length for the UE. The configured symbol length should be no less than the number of symbols occupied by PRS to ensure that the UE fully receives the PRS to be measured. Due to factors such as clock drift and UE mobility, using a symbol length equal to the number of symbols occupied by PRS may result in the UE being unable to fully receive the PRS. In this case, the symbol length can be configured to be greater than the number of symbols occupied by PRS, such as 3, 5, 7, or 13 symbols.

[0161] There are two types of measurement windows for downlink positioning: Measurement Gap (MG) and PRS Processing Window (PPW). MG is used for inter-frequency or inter-system measurements, while PPW is used for measurements within the currently activated BWP. Based on the above description, the configuration examples of MG and PPW are as follows:

[0162] When the measurement window is MG, the traditional measurement window duration is used in non-AI methods. In AI methods, the final measurement window duration should be the configured symbol length plus the receiver frequency transition duration (1ms). When the measurement window is PPW, the traditional measurement window duration is also used in non-AI methods. In AI methods, the configured symbol length is used as the measurement window duration.

[0163] Option 2: Reduce measurement time with auxiliary information

[0164] If the LMF entity 300 can obtain the position of UE 10, it can select one or more beams close to UE 10 as measurement beams of UE 10 based on the position information and indicate these beams. The number of beams is configured autonomously by the LMF entity 300 and may depend on whether the current UE 10 is located at the center of the beam or at the intersection of the beams.

[0165] The LMF entity 300 can provide DL PRS-related assistance data (Assistance Data) to the UE through LPP signaling to assist the UE 10 in measuring the PRS and calculating the position. The assistance data includes a PRS resource subset for AI positioning. The assistance data is independently configured by the LMF entity 300 and represents the PRS resource set ID and PRS resource ID corresponding to the UE 10 measurement beam corresponding to each base station 200 when the AI ​​method is adopted.

[0166] If the AI ​​model is located at the UE 10 and the UE 10 infers the location of the UE 10 through the AI ​​model, or if the AI ​​model is located at the LMF entity 300 and the LMF entity 300 infers the location of the UE 10 based on the measurement value reported by the UE 10, an example of the LPP signaling message is as follows:

[0167] If the LMF entity 300 provides assistance data for multiple positioning methods, the information element (IE) nr-DL-PRS-AssistanceData only appears in one IE among NR-Multi-RTT-ProvideAssistanceData, NR-DL-AoD-ProvideAssistanceData, NR-DL-TDOA-ProvideAssistanceData, and NR-DL-AI-ProvideAssistanceData.

[0168] If the AI ​​model is located at the UE entity 300, the UE 10 uses the model to infer intermediate measurements and reports them to the LMF entity 300 to calculate the location of the UE 10. An example of an LPP signaling message is shown below. In this case, the LMF entity 300 may configure different PRS resources for the AI ​​method and the non-AI method.

[0169] When the LMF entity 300 is configured with only one beam, that is, the cardinality of the PRS resource subset used for AI positioning is 1, the UE only measures the beam corresponding to the resource within the PRS resource group period. When the LMF entity 300 is configured with multiple beams, that is, the cardinality of the PRS resource subset used for AI positioning is greater than 1, if the AI ​​model is located on the UE side, the UE 10 can autonomously select a suitable beam of each base station 200 for model inference, such as based on measurement values ​​such as PRS-RSRP. If the AI ​​model is located on the LMF entity 300 side, the UE 10 can select one or more beams of each base station 200 (which can be beams corresponding to all resources of the PRS resource subset used for AI positioning) to report the measurement values, and the reported measurement values ​​include model input (such as PDP) and auxiliary information (such as PRS-RSRP).

[0170] In addition to the network being able to configure a shorter measurement window for UE 10 to reserve more time domain resources for sending or receiving other signals, the PRS beam scanning method is also based on manufacturer-specific implementations, and the order in which beams are transmitted may change over time. In this case, the measurement beams for UE 10 can be reconfigured by changing the resource index to maintain the measured beams unchanged. As shown in Figure 9, after the beam transmission order changes, the reference signal can be reconfigured for UE 2, and the PRS resource subset corresponding to TRP 1 for AI positioning can be configured to {1}, indicating that UE 2 can continue measuring the beam corresponding to the second PRS resource.

[0171] Downlink Positioning Model Training and Inference Consistency Method

[0172] Option 1: UE-side model

[0173] Please refer to Figure 10, which is a schematic diagram of the UE and LMF entity operating on the UE side model training and inference consistency method. When the AI ​​model is located in the UE 10, the UE trigger request method can be used.

[0174] In step 1001, the LMF entity 300 provides the UE 10 with auxiliary data related to factors affecting the consistency of model training and model inference through LPP signaling, such as the pre-configuration of on-demand PRS. This parameter indicates the PRS configuration information supported by the base station 200. Thus, an example of LPP signaling message under the AI ​​positioning mode is as follows. If the LMF entity 300 configures available on-demand PRS configurations for multiple positioning methods, the IE NR-On-Demand-DL-PRS-Configurations can only appear in one of the IEs of NR-Multi-RTT-ProvideAssistanceData, NR-DL-AoD-ProvideAssistanceData, NR-DL-TDOA-ProvideAssistanceData, and NR-DL-AI-ProvideAssistanceData.

[0175] In step 1002, UE 10 provides auxiliary information to LMF entity 300 via LPP signaling. The auxiliary information includes on-demand PRS configuration and / or cell information. The on-demand PRS configuration requests LMF entity 300 to configure base station 200 to send the required PRS to ensure consistency of PRS configuration in model training and model inference. If LMF entity 300 provides auxiliary data to UE 10, such as pre-configured on-demand PRS, UE 10 reports one or more required PRS configurations to LMF entity 300 within the pre-configured range. This may be done using either index indication or explicit indication. Index indication reports the index corresponding to the required PRS configuration, while explicit indication directly reports the PRS configuration information. If LMF entity 300 does not provide auxiliary data, UE 10 directly reports the required PRS configuration information using explicit indication. If explicit reporting is adopted, UE 10 may report at least one of the following PRS configuration parameters: frequency band, resource set period, bandwidth, repetition factor, number of symbols, comb size, QCL information, minimum number of resources per resource set or minimum power, where the introduction of the minimum number of resources takes into account that the beam size affects the multipath component. For example, a wide beam will increase the interference of adjacent multipaths, thereby affecting the accuracy of the AI ​​model. The minimum power represents the minimum average EPRE of the resource element carrying the PRS. This parameter is an integer in dBm. The cell information indicates the cell range that the LMF can configure to ensure the consistency of the cell information in model training and model inference. The cell information includes at least one of the following: PCI list, NCGI list or ARFCN list. Each list represents a set of information indicating the range that the LMF can configure.

[0176] Option 2: LMF Side Model

[0177] When the AI ​​model is located on the LMF entity 300, at least one of the following methods can be used:

[0178] 1. The LMF entity 300 configures assistance information: The LMF entity 300 provides the UE 10 with at least one of the following assistance data through LPP signaling: a PRS-RSRP threshold, a PRS-RSRQ threshold, a PRS-SINR threshold, a current model valid area, or a list of expected model valid areas. The determination of the PRS-RSRP threshold, the PRS-RSRQ threshold, and the PRS-SINR threshold is based on autonomous implementation on the network side, such as based on statistical results of measurement values ​​during model training on the network side. The current model area represents the area to which the currently activated model applies (the area is described by a cell ID list, and the cell ID includes at least one of the following: NCGI, PCI, or ARFCN). The expected model valid area list represents a set of areas (cell ID lists) to which the model applied to the surrounding area applies. Therefore, an example of an LPP signaling message related to configuring assistance information is as follows:

[0179] When the LMF entity 300 is configured with at least one of the following measurement thresholds: a PRS-RSRP threshold, a PRS-RSRQ threshold, or a PRS-SINR threshold, the UE autonomously selects the model input corresponding to one or more beams of each TRP for reporting based on the configured thresholds. The PRS-RSRP and other measurement values ​​corresponding to these beams should not be less than the configured thresholds. If the LMF entity 300 is configured with multiple measurement thresholds including the PRS-RSRP threshold, the PRS-RSRQ threshold, and the PRS-SINR threshold, the measurement values ​​corresponding to the beams used by the UE must not be less than the thresholds configured by the LMF entity 300.

[0180] When the LMF entity 300 is configured with a current model valid area, the UE 10 selects and reports the model input corresponding to the reference signal that matches the valid area (the reference signal transmitted by the cell in the cell ID list). If the LMF entity 300 is configured with a list of expected model valid areas, the UE 10 can report a region transition message to the LMF entity 300 upon entering any area in the area set (i.e., upon receiving a reference signal within the area), to assist the LMF entity 300 in model switching.

[0181] Please refer to Figure 11, which is a schematic diagram of the model training and inference consistency method operated by UE and LMF entities on the LMF side.

[0182] In step 1101, the LMF entity 300 requests measurement from the UE 10 through LPP signaling and configures the UE 10 with at least one of the following required auxiliary information: PRS-RSRP, PRS-RSRQ, PRS-SINR, or cell information. The PRS-RSRP measurement value represents the linear average of the power of the REs where the PRS is located within the measurement bandwidth, in dBm. The PRS-RSRQ measurement value is defined as the following formula:

[0183] Where N represents the number of RBs in the measurement bandwidth, NR carrier RSSI represents the linear average of the OFDM power (the power of N RBs), expressed in dB, and the PRS-SINR measurement value represents the ratio of the linear average of the power of the REs where the PRS is located to the linear average of the power of the interference signal (noise, interference) in the measurement bandwidth, expressed in dB.

[0184] In addition, the auxiliary information about the cell information is applicable to the area-specific model, which can assist the LMF entity 300 in determining the model applicable to the current area. Based on the above description, an example of the LPP signaling message is as follows:

[0185] In step 1102, if the LMF entity 300 is configured with one or more measurement quantities of PRS-RSRP, PRS-RSRQ, or PRS-SINR, the UE reports the model input corresponding to one or more beams for each TRP to the LMF entity 300 via LPP signaling, and also reports the measurement value configured by the LMF entity 300 for each beam. To reduce the reporting overhead of auxiliary information, as shown in Table 10, the following predefined PRS-RSRP reporting mapping table can be used. The PRS-RSRP range is -156 to -31 dBm, with a granularity of 1 dBm.

[0186] Table 10

[0187] In addition, a predefined PRS-RSRQ reporting mapping table and a PRS-SINR reporting mapping table can be introduced to index PRS-RSRQ and PRS-SINR, reducing the reporting overhead of the measurement quantity. Table 11 shows the PRS-RSRQ reporting mapping table and Table 12 shows the PRS-SINR reporting mapping table, where the PRS-RSRQ range is -43 to 20 dB with a granularity of 0.5 dB, and the PRS-SINR range is -23 to 40 dB with a granularity of 0.5 dB.

[0188] Table 11

[0189] Table 12

[0190] When the LMF entity 300 is configured with cell information, the UE 10 reports the model input corresponding to one or more beams of each base station to the LMF entity 300 through LPP signaling, and at the same time reports at least one of the following auxiliary information: the PRS ID of each base station 200, the PCI of each base station 200, the NCGI of each base station 200, or the ARFCN of each base station 200. Based on the above description, an example of the LPP signaling message for the UE to report auxiliary information is as follows:

[0191] Uplink positioning model training and reasoning consistency method

[0192] When the AI ​​model is located on the LMF entity 300, at least one of the following methods can be used:

[0193] 1. The LMF entity 300 configures auxiliary information: The LMF entity 300 provides the gNB with at least one of the following auxiliary information through NRPPa signaling: SRS-RSRP threshold, SRS-RSRQ threshold, SRS-SINR threshold, SRS bandwidth, SRS frequency, number of resources in an SRS resource set, period of each SRS resource, number of ports of each SRS resource, comb factor of each SRS resource, number of symbols of each SRS resource, repetition factor of each SRS resource, or SRS QCL information. The SRS-RSRP threshold, SRS-RSRQ threshold, and SRS-SINR threshold are determined autonomously by the network, such as based on statistical results of measurement values ​​during network-side model training. Therefore, an example of NRPPa signaling message related to configuring auxiliary information is as follows:

[0194] When the LMF entity 300 is configured with at least one of the following measurement thresholds: the SRS-RSRP threshold, the SRS-RSRQ threshold, or the SRS-SINR threshold, the base station 200 autonomously selects the model input corresponding to one or more beams collected from each TRP based on the configured threshold and reports it. The SRS-RSRP and other measurement values ​​corresponding to these beams must be no less than the configured threshold. If the LMF entity 300 is configured with multiple measurement thresholds, including the SRS-RSRP threshold, the SRS-RSRQ threshold, and the SRS-SINR threshold, the measurement values ​​corresponding to the beams selected by the base station 200 must be no less than the thresholds configured by the LMF entity 300.

[0195] When the LMF entity 300 is configured with at least one of the following SRS configuration information: SRS bandwidth, SRS frequency, number of resources in an SRS resource set, period of each SRS resource, number of ports of each SRS resource, comb factor of each SRS resource, number of symbols of each SRS resource, repetition factor of each SRS resource, or SRS QCL information, the base station 200 configures the SRS for the UE 10 according to the configuration information and feeds back the adopted SRS configuration to the LMF entity 300. After the LMF entity 300 instructs measurement, the base station 200 measures the SRS transmitted by the UE 10 and reports the obtained model input to the LMF entity 300.

[0196] Please refer to Figure 12, which is a schematic diagram of the base station and LMF entity operating the LMF side model training and inference consistency method.

[0197] In step 1201, the LMF entity 300 requests measurement from the gNB through NRPPa signaling and configures the gNB with at least one of the following required auxiliary information: SRS-RSRP, SRS-RSRQ, or SRS-SINR. The SRS-RSRP measurement quantity represents the linear average power of the REs where the SRS is located within the measurement bandwidth, in dBm. The SRS-RSRQ measurement quantity is defined as follows:

[0198] Where N represents the number of RBs in the measurement bandwidth, NR carrier RSSI represents the linear average of the OFDM power (the power of N RBs), expressed in dB, and the SRS-SINR measurement value represents the ratio of the linear average of the power of the REs where the SRS is located to the linear average of the power of the interference signal (noise, interference) in the measurement bandwidth, expressed in dB.

[0199] Therefore, an example NRPPa signaling message is as follows:

[0200] In step 1202, when the LMF entity 300 configures one or more measurement quantities of SRS-RSRP, SRS-RSRQ, and SRS-SINR, the base station 200 reports the model input corresponding to one or more beams collected by each base station 200 to the LMF entity 300 via NRPPa signaling, and also reports the measurement value configured by the LMF entity 300 for each beam. To reduce the reporting overhead of auxiliary information, the following predefined SRS-RSRP reporting mapping table (Table 13) can be used. The SRS-RSRP range is -156 to -31 dBm, with a granularity of 1 dBm.

[0201] Table 13

[0202] In addition, predefined SRS-RSRQ reporting mapping tables and SRS-SINR reporting mapping tables can be introduced to index SRS-RSRQ and SRS-SINR, reducing the reporting overhead of measurement quantities. Table 14 shows the SRS-RSRQ reporting mapping table, and Table 15 shows the SRS-SINR reporting mapping table. The SRS-RSRQ range is -43 to 20 dB with a granularity of 0.5 dB, and the SRS-SINR range is -23 to 40 dB with a granularity of 0.5 dB.

[0203] Table 14

[0204] Table 15

[0205] According to the above description, an example of the NRPPa signaling message for the base station 200 to report the auxiliary information is as follows:

[0206] Compared to the prior art, an embodiment of the present invention provides an artificial intelligence positioning method, which is executed in a UE and includes: receiving AI positioning assistance data and / or a model input reporting configuration from a core network device, and reporting AI positioning information based on the AI ​​positioning assistance data and / or the model input reporting configuration. Another embodiment of the present application provides an artificial intelligence positioning method, which is executed in a core network device and includes: sending AI positioning assistance data and / or a model input reporting configuration to a user equipment (UE), and receiving AI positioning information generated by the UE based on the AI ​​positioning assistance data and / or the model input reporting configuration. Another embodiment of the present application provides an artificial intelligence positioning method, which is executed in a base station and includes: receiving AI positioning assistance data and / or a model input reporting configuration from a core network device, and reporting AI positioning information based on the AI ​​positioning assistance data and / or the model input reporting configuration. Another embodiment of the present application provides an artificial intelligence positioning method, which is executed in a core network device and includes: sending AI positioning assistance data and / or a model input reporting configuration to a base station, and receiving AI positioning information generated by the base station based on the AI ​​positioning assistance data and / or the model input reporting configuration. This application can improve the efficiency of AI positioning by introducing measurement configuration and measurement quantity reporting for AI positioning. At the same time, this application can also reduce reporting signaling overhead by compressing the amount of reported data. In addition, this application can reduce resource usage by shortening the measurement time. In addition, the UE / base station and core network equipment of this application exchange auxiliary information to ensure the consistency of model training and model inference.

[0207] According to an example embodiment, a chip is provided, comprising: a processor for calling and running a computer program from a memory, so that a device equipped with the chip executes a method according to any one of the above embodiments, examples, or exemplary embodiments.

[0208] According to an example embodiment, there is provided a computer-readable storage medium for storing a computer program, wherein the computer program causes a computer to execute a method according to any one of the above-mentioned embodiments, examples, or exemplary embodiments.

[0209] According to an example embodiment, a computer program product is provided, comprising a computer program / instruction, which, when executed by a processor (e.g., by the processor or an apparatus, device, computer or machine including the processor), implements a method according to any one of the above-mentioned embodiments, examples, or example embodiments.

[0210] Embodiments of the present invention are a combination of techniques / procedures that may be employed in 3GPP specifications to create a final product.

[0211] While the present invention has been described in connection with what is considered to be the most practical and preferred embodiment, it is to be understood that the invention is not limited to the disclosed embodiment, but is intended to cover various arrangements embodied within the broadest interpretation of the appended claims.

Claims

1. A method for artificial intelligence (AI) positioning, the method being executed in a user equipment (UE), comprising: Receive AI positioning assistance data and / or model input reporting configuration from core network devices; as well as Report AI positioning information according to the AI ​​positioning assistance data and / or model input reporting configuration.

2. The method according to claim 1, wherein the AI ​​positioning assistance data comprises at least one of the following: a positioning reference signal (PRS) resource subset for AI positioning, an on-demand PRS pre-configuration, a measurement value threshold, or a model valid area.

3. The method according to claim 1, wherein when the model input reporting configuration is associated with a sampling point-based model input, the model input reporting configuration includes at least one of the following: a data type parameter, a sampling point number parameter, or a sampling granularity parameter, wherein the data type parameter is used to indicate a type of a reported channel response sequence, the sampling point number parameter is used to indicate the UE to determine the number of multiple reporting sampling points from multiple sampling points, and the sampling granularity parameter is used to indicate a screening granularity for selecting multiple reporting sampling points from the multiple sampling points.

4. The method according to claim 3, wherein the data type parameter includes at least one of the following: a delay spectrum DP, a power delay spectrum PDP, a channel impulse response CIR, a power spectrum PS or a channel frequency response CFR, the delay spectrum DP represents the delay sequence corresponding to the multiple sampling points in the channel response; the power delay spectrum PDP represents the delay and power sequence of the multiple sampling points in the channel response; the channel impulse response CIR represents the delay, power and phase sequence of the multiple sampling points in the channel response; the power spectrum PS represents the frequency and power sequence of the multiple sampling points in the channel response; and the channel frequency response CFR represents the frequency, power and phase sequence of the multiple sampling points in the channel response.

5. The method according to claim 3 or 4, wherein when the number of the sampling granularity parameter settings is p, and the data type parameter is the delay profile DP, the power delay profile PDP or the channel impulse response CIR, the UE selects the first 1 / p of the reported sampling points from the multiple sampling points.

6. The method according to claim 3 or 4, wherein when the sampling granularity parameter is set to p and the data type parameter is the power spectrum PS or the channel frequency response CFR, the UE takes the first sampling point as the starting point and selects the reporting sampling point every (p-1) sampling points.

7. The method according to claim 1, wherein the core network device is a location management function (LMF) entity.

8. The method according to claim 3, wherein the time interval between the plurality of sampling points is determined based on the bandwidth of a positioning reference signal (PRS), or the frequency interval between the plurality of sampling points is determined based on the subcarrier spacing of the PRS, and the number of the plurality of sampling points is not less than the smallest power of 2 of the number of PRS resource elements (RE).

9. The method according to claim 3, wherein the plurality of reported sampling points are represented by a bitmap and / or a sequence number of the sampling points.

10. The method according to claim 3, wherein the reporting form of the sampling values ​​of the plurality of reported sampling points comprises a bit string indication and / or a mapping table index. The method according to claim 10 , wherein the sampling values ​​of the multiple reported sampling points include power and / or power phase combination.

12. The method according to claim 3, wherein the power reporting form of each of the reported sampling points is the power difference between the current sampling and the previous sampling of each of the reported sampling points.

13. The method according to claim 3, wherein the plurality of reporting sampling points are sampling points selected from the plurality of sampling points whose power variation is greater than a power variation threshold.

14. The method according to claim 3, wherein the number of the sampling point-based model inputs to be reported is determined according to a minimum reporting number and / or a channel correlation threshold.

15. The method according to claim 1, wherein when the model input reporting configuration is associated with a path-based model input, the model input reporting configuration includes at least one of the following: a data type parameter, a multipath number parameter, a delay type parameter, or a granularity factor parameter, wherein the data type parameter is used to indicate the reported multipath parameter, the multipath number is used to indicate the number of reported multipaths, the delay type is used to indicate the reporting form of the multipath delay, and the granularity factor is used to indicate the resolution of the reported multipath delay.

16. The method according to claim 15, wherein the data type parameter includes at least one of the following: delay, delay power or delay power phase; the delay type parameter includes a relative form and / or an absolute form, the relative form represents: taking the delay of a multipath of a base station as a reference delay, the other multipath delays of the base station and the multipath delays of multiple base stations other than the base station are relative delays to the reference delay, and the absolute form is the absolute delay of the multipaths of multiple base stations.

17. The method according to claim 16, wherein the absolute delay is determined by a downlink relative arrival time (DL-RTOA) reference time of a base station and the time when the subframe containing the PRS of the base station is received, wherein the DL-RTOA reference time is T0+t PRS , T0 represents the starting time of the first system frame of the base station, t PRS Indicates the start time of the PRS relative to the first system frame.

18. The method according to claim 16, wherein the reference delay comprises the absolute delay of the first multipath of the assistance data reference base station and / or the minimum delay of all multipaths of all base stations.

19. The method according to claim 1, further comprising: A model training type configuration is received from the core network device, wherein the model training type configuration is used to indicate whether to report a positioning position and / or a positioning measurement value.

20. The method of claim 1, further comprising: Receive a measurement window duration configuration for AI positioning from a base station, wherein the measurement window duration configuration for AI positioning indicates a symbol length, and when the measurement window is associated with a measurement interval MG, the measurement window duration is the symbol length plus the receiver frequency conversion duration; when the measurement window is associated with a PRS processing window PPW, the measurement window duration is the symbol length. Spend.

21. The method according to claim 2, further comprising: receiving a pre-configuration of the on-demand PRS provided by the core network device; and According to the pre-configuration of the on-demand PRS, one or more required PRS configurations are reported to the core network device within the pre-configured range.

22. The method according to claim 1, further comprising: reporting one or more PRS configurations and / or cell information to the core network device.

23. A wireless communication device comprising: A processor and a memory, the memory being used to store a computer program, the processor being used to call and run the computer program stored in the memory to execute the method according to any one of claims 1 to 22.

24. A method for artificial intelligence (AI) positioning, the method being executed in a core network device, comprising: Send AI positioning assistance data and / or model input reporting configuration to user equipment UE; as well as Receive AI positioning information generated by the UE according to the AI ​​positioning assistance data and / or model input reporting configuration. 25 . The method according to claim 24 , wherein the AI ​​positioning assistance data comprises at least one of the following: a positioning reference signal (PRS) resource subset for AI positioning, on-demand PRS pre-configuration, a measurement value threshold, or a model valid area.

26. The method according to claim 24, wherein when the model input reporting configuration is associated with a model input based on a sampling point, the model input reporting configuration includes at least one of the following: a data type parameter, a sampling point number parameter, or a sampling granularity parameter, wherein the data type parameter is used to indicate a type of a reported channel response sequence, the sampling point number parameter is used to indicate the UE to determine the number of multiple reporting sampling points from multiple sampling points, and the sampling granularity parameter is used to indicate a screening granularity for selecting multiple reporting sampling points from the multiple sampling points.

27. The method according to claim 26, wherein the data type parameters include at least one of the following: a delay spectrum DP, a power delay spectrum PDP, a channel impulse response CIR, a power spectrum PS or a channel frequency response CFR, the delay spectrum DP represents the delay sequence corresponding to the multiple sampling points in the channel response; the power delay spectrum PDP represents the delay and power sequence of the multiple sampling points in the channel response; the channel impulse response CIR represents the delay, power and phase sequence of the multiple sampling points in the channel response; the power spectrum PS represents the frequency and power sequence of the multiple sampling points in the channel response; the channel frequency response CFR represents the frequency, power and phase sequence of the multiple sampling points in the channel response.

28. The method according to claim 26 or 27, wherein when the number of the sampling granularity parameter settings is p, and the data type parameter is the delay profile DP, the power delay profile PDP or the channel impulse response CIR, the UE selects the first 1 / p of the reported sampling points from the multiple sampling points.

29. The method according to claim 26 or 27, wherein when the sampling granularity parameter is set to p and the data type parameter is the power spectrum PS or the channel frequency response CFR, the UE takes the first sampling point as the starting point and selects the reporting sampling point every (p-1) sampling points.

30. The method according to claim 24, wherein the core network device is a Location Management Function (LMF) entity.

31. The method according to claim 26, wherein the time interval between the plurality of sampling points is determined based on the bandwidth of a positioning reference signal (PRS), or the frequency interval between the plurality of sampling points is determined based on the subcarrier spacing of the PRS, and the number of the plurality of sampling points is a minimum power of 2 that is not less than the number of PRS resource elements (RE).

32. The method according to claim 26, wherein the plurality of reported sampling points are represented by a bitmap and / or a sequence number of sampling points.

33. The method according to claim 26, wherein the reporting form of the sampling values ​​of the plurality of reported sampling points comprises a bit string indication and / or a mapping table index.

34. The method according to claim 33, wherein the sampling values ​​of the plurality of reported sampling points include power and / or power phase combination.

35. The method according to claim 26, wherein the power reporting form of each of the reported sampling points is the power difference between the current sampling and the previous sampling of each of the reported sampling points.

36. The method according to claim 26, wherein the plurality of reporting sampling points are sampling points selected from the plurality of sampling points whose power variation is greater than a power variation threshold.

37. The method of claim 26, wherein the number of the sampling point-based model inputs to be reported is determined according to a minimum reporting number and / or a channel correlation threshold.

38. The method according to claim 24, wherein when the model input reporting configuration is associated with a path-based model input, the model input reporting configuration includes at least one of the following: a data type parameter, a multipath number parameter, a delay type parameter, or a granularity factor parameter, wherein the data type parameter is used to indicate the reported multipath parameter, the multipath number is used to indicate the number of reported multipaths, the delay type is used to indicate the reporting form of the multipath delay, and the granularity factor is used to indicate the resolution of the reported multipath delay.

39. The method according to claim 38, wherein the data type parameter includes at least one of the following: delay, delay power or delay power phase; the delay type parameter includes a relative form and / or an absolute form, the relative form represents: taking the delay of a multipath of a base station as the reference delay, the other multipath delays of the base station and the multipath delays of multiple base stations other than the base station are relative delays to the reference delay, and the absolute form is the absolute delay of the multipaths of multiple base stations.

40. The method according to claim 39, wherein the absolute delay is determined by a downlink relative arrival time (DL-RTOA) reference time of a base station and a time of receiving a subframe containing a PRS from the base station, wherein the DL-RTOA reference time is T0+t PRS , T0 represents the starting time of the first system frame of the base station, t PRS Indicates the start time of the PRS relative to the first system frame.

41. The method according to claim 39, wherein the reference delay comprises an absolute delay of a first multipath of an assistance data reference base station and / or a minimum delay of all multipaths of all base stations.

42. The method of claim 24, further comprising: Sending a model training type configuration to the UE, wherein the model training type configuration is used to indicate whether to report a positioning position and / or a positioning measurement value.

43. The method of claim 25, further comprising: Sending the pre-configuration of the on-demand PRS to the UE; and Receive one or more PRS configurations reported by the UE within a pre-configured range according to the pre-configuration of the on-demand PRS.

44. The method of claim 24, further comprising: receiving one or more PRS configurations and / or cell information from the UE.

45. A wireless communication device comprising: A processor and a memory, the memory being used to store a computer program, the processor being used to call and run the computer program stored in the memory to execute the method according to any one of claims 24 to 44.

46. ​​A method for artificial intelligence (AI) positioning, the method being performed in a base station, comprising: Receive AI positioning assistance data and / or model input reporting configuration from core network devices; as well as Report AI positioning information according to the AI ​​positioning assistance data and / or model input reporting configuration. The method according to claim 46 , wherein the AI ​​positioning assistance data comprises a sounding reference signal (SRS) configuration and / or a measurement quantity threshold.

48. The method according to claim 46, wherein when the model input reporting configuration is associated with a model input based on a sampling point, the model input reporting configuration includes at least one of the following: a data type parameter, a sampling point number parameter, or a sampling granularity parameter, the data type parameter is used to indicate the type of the reported channel response sequence, the sampling point number parameter is used to indicate the UE to determine the number of multiple reporting sampling points from multiple sampling points, and the sampling granularity parameter is used to indicate the screening granularity used to select multiple reporting sampling points from the multiple sampling points.

49. The method according to claim 48, wherein the data type parameters include at least one of the following: a delay spectrum DP, a power delay spectrum PDP, a channel impulse response CIR, a power spectrum PS or a channel frequency response CFR, the delay spectrum DP represents the delay sequence corresponding to the multiple sampling points in the channel response; the power delay spectrum PDP represents the delay and power sequence of the multiple sampling points in the channel response; the channel impulse response CIR represents the delay, power and phase sequence of the multiple sampling points in the channel response, the power spectrum PS represents the frequency and power sequence of the multiple sampling points in the channel response, and the channel frequency response CFR represents the frequency, power and phase sequence of the multiple sampling points in the channel response.

50. The method according to claim 48 or 49, wherein when the number of the sampling granularity parameter settings is p, and the data type parameter is the delay profile DP, the power delay profile PDP or the channel impulse response CIR, the UE selects the first 1 / p of the reported sampling points from the multiple sampling points.

51. The method according to claim 48 or 49, wherein when the sampling granularity parameter is set to p and the data type parameter is the power spectrum PS or the channel frequency response CFR, the UE takes the first sampling point as the starting point and selects the reporting sampling point every (p-1) sampling points.

52. The method according to claim 46, wherein the core network device is a Location Management Function (LMF) entity.

53. The method according to claim 48, wherein the time interval between the multiple sampling points is determined based on the bandwidth of the positioning reference signal (SRS), or the frequency interval between the multiple sampling points is determined based on the subcarrier spacing of the SRS, and the number of the multiple sampling points is not less than the smallest power of 2 of the number of SRS resource elements (RE).

54. The method according to claim 48, wherein the plurality of reported sampling points are represented by a bitmap and / or a sequence number of sampling points.

55. The method according to claim 48, wherein the reporting form of the sampling values ​​of the plurality of reported sampling points comprises a bit string indication and / or a mapping table index.

56. The method according to claim 55, wherein the sampling values ​​of the plurality of reported sampling points include power and / or power phase combination.

57. The method according to claim 48, wherein the power reporting form of each of the reported sampling points is the power difference between the current sampling and the previous sampling of each of the reported sampling points.

58. The method according to claim 48, wherein the plurality of reporting sampling points are sampling points selected from the plurality of sampling points whose power variation is greater than a power variation threshold.

59. The method of claim 48, wherein the number of the sampling point-based model inputs to be reported is determined according to a minimum reporting number and / or a channel correlation threshold.

60. The method according to claim 46, wherein when the model input reporting configuration is associated with a path-based model input, the model input reporting configuration includes at least one of the following: a data type parameter, a multipath number parameter, or a granularity factor parameter, the data type parameter is used to indicate the reported multipath parameter, the multipath number is used to indicate the number of reported multipaths, and the granularity factor is used to indicate the resolution of the reported multipath delay.

61. The method of claim 48, wherein the data type parameter comprises at least one of: delay, delay power, or delay power phase.

62. The method of claim 46, further comprising: A model training type configuration is received from the core network device, wherein the model training type configuration is used to indicate whether to report a positioning measurement value.

63. A method for artificial intelligence (AI) positioning, the method being executed in a core network device, comprising: Send AI positioning assistance data and / or model input reporting configuration to the base station; as well as Receive AI positioning information generated by the base station according to the AI ​​positioning assistance data and / or model input reporting configuration. The method according to claim 63 , wherein the AI ​​positioning assistance data comprises a sounding reference signal (SRS) configuration and / or a measurement quantity threshold.

65. The method according to claim 63, wherein when the model input reporting configuration is associated with a model input based on a sampling point, the model input reporting configuration includes at least one of the following: a data type parameter, a sampling point number parameter, or a sampling granularity parameter, the data type parameter is used to indicate a type of a reported channel response sequence, the sampling point number parameter is used to indicate the UE to determine the number of multiple reporting sampling points from multiple sampling points, and the sampling granularity parameter is used to indicate a screening granularity for selecting multiple reporting sampling points from the multiple sampling points.

66. A method according to claim 65, wherein the data type parameters include at least one of the following: a delay spectrum DP, a power delay spectrum PDP, a channel impulse response CIR, a power spectrum PS or a channel frequency response CFR, the delay spectrum DP represents the delay sequence corresponding to the multiple sampling points in the channel response; the power delay spectrum PDP represents the delay and power sequence of the multiple sampling points in the channel response; the channel impulse response CIR represents the delay, power and phase sequence of the multiple sampling points in the channel response, the power spectrum PS represents the frequency and power sequence of the multiple sampling points in the channel response, and the channel frequency response CFR represents the frequency, power and phase sequence of the multiple sampling points in the channel response.

67. The method according to claim 65 or 66, wherein when the number of the sampling granularity parameter settings is p, and the data type parameter is the delay profile DP, the power delay profile PDP or the channel impulse response CIR, the UE selects the first 1 / p of the reported sampling points from the multiple sampling points.

68. The method according to claim 65 or 66, wherein when the sampling granularity parameter is set to p and the data type parameter is the power spectrum PS or the channel frequency response CFR, the UE takes the first sampling point as the starting point and selects the reporting sampling point every p-1 sampling points.

69. The method according to claim 63, wherein the core network device is a Location Management Function (LMF) entity.

70. The method according to claim 65, wherein the time interval between the multiple sampling points is determined based on the bandwidth of the positioning reference signal SRS, or the frequency interval between the multiple sampling points is determined based on the subcarrier spacing of the SRS, and the number of the multiple sampling points is not less than the smallest power of 2 of the number of SRS resource elements RE.

71. The method according to claim 65, wherein the plurality of reported sampling points are represented by a bitmap and / or a sequence number of sampling points.

72. The method according to claim 65, wherein the reporting form of the sampling values ​​of the plurality of reported sampling points comprises a bit string indication and / or a mapping table index.

73. The method according to claim 72, wherein the sampling values ​​of the plurality of reported sampling points include power and / or power phase combination.

74. The method according to claim 65, wherein the power reporting form of each of the reported sampling points is the power difference between the current sampling and the previous sampling of each of the reported sampling points.

75. The method according to claim 65, wherein the plurality of reporting sampling points are sampling points selected from the plurality of sampling points whose power variation is greater than a power variation threshold.

76. The method of claim 65, wherein the number of the sampling point-based model inputs to be reported is determined according to a minimum reporting number and / or a channel correlation threshold.

77. The method according to claim 63, wherein when the model input reporting configuration is associated with a path-based model input, the model input reporting configuration includes at least one of the following: a data type parameter, a multipath number parameter, and a granularity factor parameter, wherein the data type parameter is used to indicate the reported multipath parameter, the multipath number is used to indicate the number of reported multipaths, and the granularity factor is used to indicate the resolution of the reported multipath delay.

78. The method of claim 65, wherein the data type parameter comprises at least one of: delay, delay power, or delay power phase.

79. The method of claim 63, further comprising: Sending a model training type configuration to the base station, wherein the model training type configuration is used to indicate whether to report the positioning measurement value.

80. A wireless communication device comprising: A processor and a memory, the memory being used to store a computer program, the processor being used to call and run the computer program stored in the memory to execute the method as claimed in any one of claims 46 to 62.

81. A wireless communication device comprising: A processor and a memory, the memory being used to store a computer program, the processor being used to call and run the computer program stored in the memory to execute the method as claimed in any one of claims 63 to 79.

Images