Terminal positioning method and device, computer device, storage medium and program product

Abstract

The application relates to a terminal positioning method and device, computer equipment, a storage medium and a program product. The fingerprint information of a service base station measured by a plurality of terminals and the distances between the plurality of terminals are acquired, the relationship graph structure data between the terminals is determined according to the fingerprint information and the distances between the plurality of terminals, the predicted positions of the terminals are determined according to the relationship graph structure data and a graph attention network, and thus the terminal positions are positioned. In the embodiment, a high-density UWB base station does not need to be deployed in a scene, and thus the terminal positioning cost can be saved.

Description

Technical Field

[0001] This application relates to the field of terminal positioning technology, and in particular to a terminal positioning method, apparatus, computer equipment, storage medium, and program product. Background Technology

[0002] With the continuous advancement of positioning technology, it has been widely applied to various intelligent business scenarios, such as metaverse and autonomous driving.

[0003] Currently, ultra-wideband (UWB) technology is used for positioning. UWB technology is a high-precision radio frequency positioning technology that has emerged in recent years. It achieves ultra-high temporal resolution through ultra-high bandwidth, thereby achieving centimeter-level positioning results.

[0004] However, high-density UWB base stations need to be deployed in advance in the scenario to achieve good accuracy and coverage, which is very costly. Summary of the Invention

[0005] Therefore, it is necessary to provide a terminal positioning method, device, computer equipment, storage medium, and program product that can reduce deployment costs in response to the above-mentioned technical problems.

[0006] Firstly, this application provides a terminal positioning method. The method includes:

[0007] Obtain fingerprint information of the serving base station measured by multiple terminals, as well as the distance between these multiple terminals;

[0008] Based on the fingerprint information and the distance between the multiple terminals, determine the relationship graph structure data between each terminal;

[0009] Based on the relational graph structure data and the graph attention network, the predicted location of each terminal is determined.

[0010] In one embodiment, step A includes: the distance is determined by the terminal based on ultra-wideband signals received from other terminals among the plurality of terminals.

[0011] In one embodiment, the distance is determined when the difference between the total received power of the terminal and the signal power of the first received ultra-wideband signal is greater than a preset threshold; the total received power is determined based on the channel impulse response power, preamble accumulation count, and impulse repetition frequency of the ultra-wideband signal.

[0012] In one embodiment, the method further includes:

[0013] Obtain multiple sample relationship graph structure data; this sample relationship graph structure data is determined based on the sample fingerprint information of the serving base station measured by multiple terminals and the sample distance between multiple terminals;

[0014] The initial graph attention network is trained based on the graph structure data of the multiple sample relationships, and the graph attention network is obtained.

[0015] In one embodiment, training an initial graph attention network based on the plurality of sample relationship graph structure data to obtain the graph attention network includes:

[0016] Input the structural data of the relationship graph of each sample into the initial graph attention network to obtain the predicted location label of each terminal;

[0017] Based on the predicted location label for each terminal, the location coordinates corresponding to the predicted location label, the actual location label, and the location coordinates corresponding to the actual location label, the initial graph attention network is trained to obtain the graph attention network.

[0018] In one embodiment, determining the relationship graph structure data between the terminals based on the fingerprint information and the distance between the multiple terminals includes:

[0019] Each fingerprint information is used as node feature information;

[0020] The distances between these multiple terminals are used as edge information;

[0021] Based on the node's feature information and edge information, determine the relational graph structure data between each terminal.

[0022] In one embodiment, the fingerprint information includes at least one of a reference signal received power and a received signal strength indication.

[0023] Secondly, this application also provides a terminal positioning device. The device includes:

[0024] The acquisition module is used to acquire fingerprint information of the serving base station measured by multiple terminals, as well as the distance between the multiple terminals;

[0025] The first determining module is used to determine the relationship graph structure data between the terminals based on the fingerprint information and the distance between the multiple terminals;

[0026] The second determining module is used to determine the predicted position of each terminal based on the relational graph structure data and the graph attention network.

[0027] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the above-described method.

[0028] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the steps of the above-described method.

[0029] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the steps of the above-described method.

[0030] The aforementioned terminal positioning method, apparatus, computer equipment, storage medium, and program product acquire fingerprint information of serving base stations measured by multiple terminals, as well as the distances between the multiple terminals. Based on the fingerprint information and the distances between the multiple terminals, they determine the relationship graph structure data between the terminals. Based on the relationship graph structure data and a graph attention network, they determine the predicted location of each terminal, thereby achieving terminal positioning. Since this embodiment does not require the deployment of a high density of UWB base stations in the scene, it can save terminal positioning costs. Attached Figure Description

[0031] Figure 1 This is an internal structural diagram of a computer device in one embodiment;

[0032] Figure 2 This is a flowchart illustrating a terminal positioning method provided in an embodiment of this application;

[0033] Figure 3 This is a schematic diagram of a multi-terminal relationship provided in an embodiment of this application;

[0034] Figure 4 This is a schematic flowchart of a model training method provided in an embodiment of this application;

[0035] Figure 5 This is a flowchart illustrating another model training method provided in an embodiment of this application;

[0036] Figure 6 This is a flowchart illustrating a method for determining relational graph structure data according to an embodiment of this application;

[0037] Figure 7 This is a schematic diagram of a terminal positioning process provided in an embodiment of this application;

[0038] Figure 8 This is a structural block diagram of a terminal positioning device provided in an embodiment of this application. Detailed Implementation

[0039] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0040] The terminal positioning method provided in this application embodiment can be applied to, for example, Figure 1 The computer device shown can be a server, and its internal structure diagram can be as follows: Figure 1 As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores relational graph structure data. The network interface communicates with external terminals via a network connection. When executed by the processor, the computer program implements a terminal location method.

[0041] Those skilled in the art will understand that Figure 1 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0042] In one embodiment, such as Figure 2 As shown, Figure 2 This is a flowchart illustrating a terminal positioning method provided in an embodiment of this application, applied to... Figure 1 Taking a computer device as an example, the explanation includes the following steps:

[0043] S201. Obtain the fingerprint information of the serving base station measured by multiple terminals, as well as the distance between multiple terminals.

[0044] The terminal can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, etc. Portable wearable devices can include smartwatches, smart bracelets, head-mounted devices, etc.

[0045] Fingerprint information may include Reference Signal Receiving Power (RSRP), Received Signal Strength Indication (RSSI), or both RSRP and RSSI. This fingerprint information is E-CID fingerprint information, where E-CID stands for Enhanced Cell-ID, an enhanced positioning technology based on Cell ID. Distance between terminals can be measured using near-field communication (NFC) technologies, such as Bluetooth and NFC. Distance between terminals can also be measured using UWB technology.

[0046] like Figure 3 As shown, Figure 3 This is a schematic diagram of a multi-terminal relationship provided in an embodiment of this application. For example, for terminal 301, the distance between terminal 301 and terminals 302 and 304 is obtained; for terminal 302, the distance between terminal 302 and terminals 301, 303, 304, and 305 is obtained; for terminal 303, the distance between terminal 303 and terminals 302 and 305 is obtained; for terminal 304, the distance between terminal 304 and terminals 301, 302, and 305 is obtained; and for terminal 305, the distance between terminal 305 and terminals 302, 303, and 304 is obtained. Each terminal can determine its distance from other terminals and send the determined distance to a computer device, thereby enabling the computer device to obtain the distances between the terminals. Each terminal can also measure the fingerprint information of the serving base station 306.

[0047] S202. Based on the fingerprint information and the distance between multiple terminals, determine the relationship graph structure data between each terminal.

[0048] In this embodiment, the relationship graph structure data is multi-user relationship graph structure data, which can use fingerprint information as node feature information and the distance between multiple terminals as edge information to construct the relationship graph structure data.

[0049] S203. Based on the relational graph structure data and the graph attention network, determine the predicted location of each terminal.

[0050] Graph attention networks can input relational graph structure data and output the predicted positions of each terminal. Graph attention networks are neural networks based on attention mechanisms.

[0051] The terminal positioning method provided in this application acquires fingerprint information of serving base stations measured by multiple terminals, as well as the distances between the multiple terminals. Based on the fingerprint information and the distances between the multiple terminals, it determines the relationship graph structure data between the terminals. Based on the relationship graph structure data and a graph attention network, it determines the predicted location of each terminal, thereby achieving terminal positioning. Since this embodiment does not require the deployment of a high density of UWB base stations in the scene, it can save terminal positioning costs.

[0052] In one embodiment, the distance is determined by the terminal based on ultra-wideband signals received from other terminals among a plurality of terminals.

[0053] The terminal can determine the time of flight of the received ultra-wideband signal and multiply the time of flight by the speed of light to obtain the distance between the terminal and other terminals. For example, combining... Figure 3 As shown, if the flight time of the ultra-wideband signal sent by terminal 302 to terminal 301 is t seconds, then the product of the speed of light and t can be used as the distance between terminal 301 and terminal 302.

[0054] Ultra-wideband (UWB) technology is a high-precision radio frequency (RF) positioning technology that has emerged in recent years. It achieves ultra-high temporal resolution through ultra-high bandwidth, resulting in centimeter-level indoor positioning accuracy. While UWB offers high positioning accuracy, it requires the deployment of a high density of UWB base stations in the environment to achieve good accuracy and coverage, leading to high deployment costs. However, with more and more terminals having built-in UWB functionality, only terminals supporting UWB are needed to determine the distance between them based on received UWB signals. This distance is then used as edge information to construct a relational graph structure, which is used for terminal positioning. This eliminates the need to deploy multiple UWB base stations, saving on infrastructure deployment costs. Furthermore, UWB's high ranging accuracy allows for the construction of a more accurate relational graph structure, improving its accuracy and further enhancing terminal positioning accuracy without requiring additional UWB base station deployments.

[0055] In one embodiment, the distance is determined when the difference between the terminal's total received power and the signal power of the first received ultra-wideband signal is less than a preset threshold; the total received power is determined based on the channel impulse response power, preamble accumulation count, and impulse repetition frequency of the ultra-wideband signal.

[0056] It should be noted that, in this embodiment of the application, to further improve the accuracy of the measured distance, before calculating the distance, it can be determined whether the ultra-wideband signal is a signal in a LOS scenario. A LOS scenario refers to a communication environment without obstructions. A LOS scenario refers to a communication environment where there is a straight transmission path between the transmitter and receiver, without any obstructions. In a LOS scenario, the signal transmission distance is longer, the signal strength is relatively higher, and the signal-to-noise ratio is better, providing more stable and reliable communication quality. However, in a NLOS scenario, the signal passes through obstacles during its journey from the transmitter to the receiver, causing attenuation and reflection, which affects the time of signal reception. Therefore, the distance calculated in a NLOS scenario will have a certain error compared to that in a LOS scenario.

[0057] In this embodiment, the LOS or NLOS scenario can be determined by calculating the power of the received signal. For example, if the difference between the total received power of the terminal and the signal power of the first received ultra-wideband signal is less than a preset threshold, it can be determined as a LOS scenario, and the distance is then calculated in the LOS scenario.

[0058] The total received power of the terminal can be calculated using the following formula:

[0059]

[0060] The signal power of the first received ultra-wideband (UWB) signal is denoted by PFP, which refers to the UWB signal arriving for the first time (First Path, FP). The unit is dBm. The signal power of the first received UWB signal can be calculated using the following formula:

[0061]

[0062] Here, F1, F2, and F3 represent the amplitudes at three points on the first arrival of the ultra-wideband signal. C, N, and A represent the channel impulse response power, the preamble accumulation count, and the pulse repetition frequency, respectively, and these values ​​can be obtained from the interface provided by the terminal's UWB chip. F1, F2, and F3 are the outputs of the UWB chip on the signal measurement results.

[0063] Optionally, the ratio of the difference between PRX and PFP to PRX can also be determined. If the ratio is less than a preset ratio, it is determined to be a LOS scene. That is, the distance can be the distance determined when the ratio of the difference between PRX and PFP to PRX is less than the preset ratio.

[0064] The method provided in this application determines the distance when the difference between the total received power of the terminal and the signal power of the first received ultra-wideband signal is less than a preset threshold, which can further improve the accuracy of the determined distance.

[0065] In one embodiment, such as Figure 4 As shown, Figure 4 This is a flowchart illustrating a model training method provided in an embodiment of this application. The method may include the following steps:

[0066] S401. Obtain multiple sample relationship graph structure data; the sample relationship graph structure data is determined based on the sample fingerprint information of the serving base station measured by multiple terminals and the sample distance between multiple terminals.

[0067] Sample fingerprint information can be used as sample node feature information, and the sample distance between multiple terminals can be used as sample edge information. Sample relationship graph structure data can be constructed based on sample node feature information and sample edge information. The constructed sample relationship graph structure data can be stored in a database, and the sample relationship graph structure data can be directly obtained from the database during the training phase.

[0068] S402. Train an initial graph attention network based on the structure data of multiple sample relationship graphs to obtain the graph attention network.

[0069] The method provided in this embodiment obtains sample relationship graph structure data determined by sample fingerprint information of serving base stations measured by multiple terminals and sample distances between multiple terminals, and trains an initial graph attention network based on the sample relationship graph structure data to obtain a graph attention network. This enables the positioning of terminals through the graph attention network, thereby reducing positioning costs.

[0070] In one embodiment, such as Figure 5 As shown, Figure 5 This is a flowchart illustrating another model training method provided in this application embodiment. This embodiment relates to a possible implementation of how to train an initial graph attention network based on multiple sample relationship graph structure data to obtain the graph attention network. Based on the above embodiment, S402 may include the following steps:

[0071] S501. Input the structural data of the relationship graph of each sample into the initial graph attention network to obtain the predicted location label corresponding to each terminal.

[0072] S502. Based on the predicted location label, the location coordinates corresponding to the predicted location label, the real location label, and the location coordinates corresponding to the real location label for each terminal, train the initial graph attention network to obtain the graph attention network.

[0073] The loss function can be used to calculate the loss value, and the network parameters can be adjusted based on the loss value to obtain a graph attention network. The loss function is shown in the following formula:

[0074]

[0075] Where ε represents the loss function, y i p represents the actual location label. i represents the predicted location label, l' represents the location coordinates corresponding to the actual location label, and l represents the location coordinates corresponding to the actual location label.

[0076] The method provided in this embodiment obtains the predicted location labels corresponding to each terminal by inputting the structural data of the relationship graph of each sample into an initial graph attention network. Based on the predicted location labels, the location coordinates corresponding to the predicted location labels, the real location labels, and the location coordinates corresponding to the real location labels, the initial graph attention network is trained to obtain a graph attention network. This enables the positioning of terminals through the graph attention network, thereby reducing positioning costs.

[0077] In one embodiment, such as Figure 6 As shown, Figure 6 This is a flowchart illustrating a method for determining relational graph structure data according to an embodiment of this application. S202 described above may include the following steps:

[0078] S601. Use each fingerprint information as node feature information.

[0079] S602, Use the distance between multiple terminals as edge information.

[0080] S603. Based on node feature information and edge information, determine the relationship graph structure data between each terminal.

[0081] Node feature information refers to node attributes, and edge information refers to edge attributes. Based on node attributes and edge attributes, the relationship graph structure data between each terminal can be determined.

[0082] like Figure 7 As shown, Figure 7 This is a schematic diagram of a terminal positioning process provided in an embodiment of this application. For example, Figure 7 The relational graph structure data 701 is input into the graph attention network. The graph attention network extracts collaborative information, and the extracted information is processed by Softmax to output the final predicted position.

[0083] Figure 7The diagram shows users 1, 2, 3, and 4, with one user corresponding to one terminal. For example, user 1 corresponds to terminal 1, user 2 to terminal 2, user 3 to terminal 3, and user 4 to terminal 4. Terminal 1 measures the distances to terminals 2, 3, and 4; terminal 2 measures the distances to terminals 1 and 3; terminal 3 measures the distances to terminals 1 and 2; and terminal 4 measures the distances to terminals 1 and 3. The constructed relational graph structure is as follows: Figure 8 The diagram shows the relationship between node feature information x1, x2, x3, and x4 corresponding to terminal 1. x1', x2', x3', and x4' are the new node feature information corresponding to each terminal, representing the output of the intermediate layer of the graph attention network. Predicted position 1, predicted position 2, predicted position 3, and predicted position 4 are the predicted positions of terminals 1, 2, 3, and 4, respectively.

[0084] To provide a clearer description of the embodiments of this application, a system flow provided by this application embodiment is described herein. This system flow includes an offline stage and an online stage. The offline stage is the stage for training an initial attention network. Based on the established fingerprint database and the distance between terminals, a sample relationship graph structure data is constructed. This sample relationship graph structure data is used to train the initial graph attention network to obtain the graph attention network; that is, the graph attention network is the trained network. In the online stage, fingerprint information from multiple terminals is acquired, and after the terminals acquire UWB signals, the distance between terminals is determined based on the UWB signals. The fingerprint information is used as node feature information, and the distance is used as edge information to construct relationship graph structure data. This relationship graph structure data is input into the graph attention network to obtain the predicted location of the terminal, i.e., the terminal's localization result, and the localization result is returned to the terminal.

[0085] It should be noted that as the number of collaborators increases, the location feature information associated with multiple users becomes richer, the corresponding location error gradually decreases, and the location accuracy will be further improved.

[0086] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0087] Based on the same inventive concept, this application also provides a terminal positioning device for implementing the terminal positioning method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more terminal positioning device embodiments provided below can be found in the limitations of the terminal positioning method described above, and will not be repeated here.

[0088] In one embodiment, such as Figure 8 As shown, Figure 8 This is a structural block diagram of a terminal positioning device provided in an embodiment of this application. The device 800 includes:

[0089] The first acquisition module 801 is used to acquire fingerprint information of the serving base station measured by multiple terminals, as well as the distance between the multiple terminals;

[0090] The first determining module 802 is used to determine the relationship graph structure data between the terminals based on the fingerprint information and the distance between the multiple terminals;

[0091] The second determining module 803 is used to determine the predicted position of each terminal based on the relational graph structure data and the graph attention network.

[0092] In one embodiment, the distance is determined by the terminal based on ultra-wideband signals received from other terminals among a plurality of terminals.

[0093] In one embodiment, the distance is determined when the difference between the terminal's total received power and the signal power of the first received ultra-wideband signal is greater than a preset threshold; the total received power is determined based on the channel impulse response power of the ultra-wideband signal, the preamble accumulation count, and the impulse repetition frequency.

[0094] In one embodiment, the device 800 may further include:

[0095] The second acquisition module is used to acquire multiple sample relationship graph structure data; the sample relationship graph structure data is determined based on the sample fingerprint information of the serving base station measured by multiple terminals and the sample distance between multiple terminals;

[0096] The module is used to train an initial graph attention network based on the graph structure data of multiple sample relationships, and then obtain the graph attention network.

[0097] In one embodiment, the obtaining module is specifically used to input the structural data of each sample relationship graph into the initial graph attention network to obtain the predicted position label corresponding to each terminal; and to train the initial graph attention network based on the predicted position label corresponding to each terminal, the position coordinates corresponding to the predicted position label, the real position label, and the position coordinates corresponding to the real position label to obtain the graph attention network.

[0098] In one embodiment, the first determining module 802 is specifically used to use each fingerprint information as node feature information; use the distance between multiple terminals as edge information; and determine the relationship graph structure data between each terminal based on the node feature information and the edge information.

[0099] In one embodiment, the fingerprint information includes at least one of a reference signal received power and a received signal strength indication.

[0100] Each module in the aforementioned terminal positioning device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0101] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0102] Obtain fingerprint information of the serving base station measured by multiple terminals, as well as the distance between multiple terminals;

[0103] Based on fingerprint information and the distances between multiple terminals, determine the relationship graph structure data between the terminals;

[0104] The predicted location of each terminal is determined based on the relational graph structure data and the graph attention network. In one embodiment, the distance is determined by the terminal based on the ultra-wideband signals received from other terminals among a plurality of terminals.

[0105] In one embodiment, the distance is determined when the difference between the terminal's total received power and the signal power of the first received ultra-wideband signal is greater than a preset threshold; the total received power is determined based on the channel impulse response power, preamble accumulation count, and impulse repetition frequency of the ultra-wideband signal.

[0106] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0107] Obtain multiple sample relationship graph structure data; the sample relationship graph structure data is determined based on the sample fingerprint information of the serving base station measured by multiple terminals and the sample distance between multiple terminals;

[0108] An initial graph attention network is trained based on multiple sample relationship graph structure data to obtain the graph attention network.

[0109] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0110] Input the structural data of the relationship graph of each sample into the initial graph attention network to obtain the predicted position label of each terminal; based on the predicted position label of each terminal, the position coordinates corresponding to the predicted position label, the real position label, and the position coordinates corresponding to the real position label, train the initial graph attention network to obtain the graph attention network.

[0111] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0112] Each fingerprint is used as a node feature; the distance between multiple terminals is used as an edge; and the relationship graph structure data between the terminals is determined based on the node feature and edge information.

[0113] In one embodiment, the fingerprint information includes at least one of a reference signal received power and a received signal strength indication.

[0114] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0115] Obtain fingerprint information of the serving base station measured by multiple terminals, as well as the distance between multiple terminals;

[0116] Based on fingerprint information and the distances between multiple terminals, determine the relationship graph structure data between the terminals;

[0117] The predicted location of each terminal is determined based on the relational graph structure data and the graph attention network. In one embodiment, the distance is determined by the terminal based on the ultra-wideband signals received from other terminals among a plurality of terminals.

[0118] In one embodiment, the distance is determined when the difference between the terminal's total received power and the signal power of the first received ultra-wideband signal is greater than a preset threshold; the total received power is determined based on the channel impulse response power, preamble accumulation count, and impulse repetition frequency of the ultra-wideband signal.

[0119] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0120] Obtain multiple sample relationship graph structure data; the sample relationship graph structure data is determined based on the sample fingerprint information of the serving base station measured by multiple terminals and the sample distance between multiple terminals;

[0121] An initial graph attention network is trained based on multiple sample relationship graph structure data to obtain the graph attention network.

[0122] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0123] Input the structural data of the relationship graph of each sample into the initial graph attention network to obtain the predicted position label of each terminal; based on the predicted position label of each terminal, the position coordinates corresponding to the predicted position label, the real position label, and the position coordinates corresponding to the real position label, train the initial graph attention network to obtain the graph attention network.

[0124] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0125] Each fingerprint is used as a node feature; the distance between multiple terminals is used as an edge; and the relationship graph structure data between the terminals is determined based on the node feature and edge information.

[0126] In one embodiment, the fingerprint information includes at least one of a reference signal received power and a received signal strength indication.

[0127] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0128] Obtain fingerprint information of the serving base station measured by multiple terminals, as well as the distance between multiple terminals;

[0129] Based on fingerprint information and the distances between multiple terminals, determine the relationship graph structure data between the terminals;

[0130] The predicted location of each terminal is determined based on the relational graph structure data and the graph attention network. In one embodiment, the distance is determined by the terminal based on the ultra-wideband signals received from other terminals among a plurality of terminals.

[0131] In one embodiment, the distance is determined when the difference between the terminal's total received power and the signal power of the first received ultra-wideband signal is greater than a preset threshold; the total received power is determined based on the channel impulse response power, preamble accumulation count, and impulse repetition frequency of the ultra-wideband signal.

[0132] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0133] Obtain multiple sample relationship graph structure data; the sample relationship graph structure data is determined based on the sample fingerprint information of the serving base station measured by multiple terminals and the sample distance between multiple terminals;

[0134] An initial graph attention network is trained based on multiple sample relationship graph structure data to obtain the graph attention network.

[0135] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0136] Input the structural data of the relationship graph of each sample into the initial graph attention network to obtain the predicted position label of each terminal; based on the predicted position label of each terminal, the position coordinates corresponding to the predicted position label, the real position label, and the position coordinates corresponding to the real position label, train the initial graph attention network to obtain the graph attention network.

[0137] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0138] Each fingerprint is used as a node feature; the distance between multiple terminals is used as an edge; and the relationship graph structure data between the terminals is determined based on the node feature and edge information.

[0139] In one embodiment, the fingerprint information includes at least one of a reference signal received power and a received signal strength indication.

[0140] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0141] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0142] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0143] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A terminal positioning method, characterized in that, The method includes: The fingerprint information of the serving base station measured by multiple terminals, as well as the distance between the multiple terminals, are obtained. The distance is determined when the difference between the total received power of the terminal and the signal power of the first received ultra-wideband signal is greater than a preset threshold. The total received power is determined based on the channel impulse response power, preamble accumulation count, and impulse repetition frequency of the ultra-wideband signal. Based on the fingerprint information and the distances between the multiple terminals, determine the relationship graph structure data between each of the terminals; Based on the relational graph structure data and the graph attention network, the predicted position of each terminal is determined; the graph attention network is a neural network based on an attention mechanism. The step of determining the relationship graph structure data between the terminals based on the fingerprint information and the distance between the multiple terminals includes: Each fingerprint information is used as node feature information; The distance between the multiple terminals is used as edge information; Based on the node feature information and edge information, determine the relationship graph structure data between the terminals; The method further includes: Multiple sample relationship graph structure data are obtained; the sample relationship graph structure data is determined based on the sample fingerprint information of the serving base station measured by multiple terminals and the sample distance between multiple terminals; Input the sample relationship graph structure data into the initial graph attention network to obtain the predicted location label for each terminal; The initial graph attention network is trained based on the predicted location labels corresponding to each terminal, the location coordinates corresponding to the predicted location labels, the actual location labels, and the location coordinates corresponding to the actual location labels, to obtain the graph attention network.

2. The method according to claim 1, characterized in that, The fingerprint information includes at least one of reference signal received power and received signal strength indication.

3. A terminal positioning device, characterized in that, The device includes: The first acquisition module is used to acquire fingerprint information of the serving base station measured by multiple terminals, as well as the distance between the multiple terminals; the distance is determined when the difference between the total received power of the terminal and the signal power of the first received ultra-wideband signal is greater than a preset threshold; the total received power is determined based on the channel impulse response power, preamble accumulation count, and impulse repetition frequency of the ultra-wideband signal. The first determining module is used to determine the relationship graph structure data between the terminals based on the fingerprint information and the distance between the multiple terminals; The second determining module is used to determine the predicted position of each terminal based on the relation graph structure data and the graph attention network; the graph attention network is a neural network based on an attention mechanism. The step of determining the relationship graph structure data between the terminals based on the fingerprint information and the distance between the multiple terminals includes: Each fingerprint information is used as node feature information; The distance between the multiple terminals is used as edge information; Based on the node feature information and edge information, determine the relationship graph structure data between the terminals; The method further includes: Multiple sample relationship graph structure data are obtained; the sample relationship graph structure data is determined based on the sample fingerprint information of the serving base station measured by multiple terminals and the sample distance between multiple terminals; Input the sample relationship graph structure data into the initial graph attention network to obtain the predicted location label for each terminal; The initial graph attention network is trained based on the predicted location labels corresponding to each terminal, the location coordinates corresponding to the predicted location labels, the actual location labels, and the location coordinates corresponding to the actual location labels, to obtain the graph attention network.

4. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 2.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 2.

6. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 2.

Images