Training method and apparatus for positioning models, method and apparatus for positioning

Abstract

This application provides a method for training a positioning model, a method for positioning, a training device for a positioning model, and a device for positioning. The method for training a positioning model includes the steps of acquiring a first channel impulse response (CIR), the first CIR being obtained based on measurement of a reference signal of a first reference point, integrating the first CIR and the first position tag to form first tag data, and training the positioning model based on the first tag data, wherein the reference signal is used for positioning, and the reference signal is transmitted or received by one or more communication devices, including a first communication device. In a 5G communication system, the CIR of the reference signal is easily collected. Therefore, a large amount of CIR data can be acquired. Based on a large amount of CIR data, a more accurate positioning model can be obtained through training. That is, when positioning is performed using the positioning model, the accuracy of the obtained positioning result is greatly improved.

Description

[Technical Field]

[0001] This application relates to the field of communications technology, and more specifically to a method for training a positioning model, a method for positioning, a training device for a positioning model, and a device for positioning. [Background technology]

[0002] With the advancement of communication technology, 5G network equipment is being deployed in large numbers, ushering in the 5G-A era. In many environments, terminal devices can easily access the 5G network. Achieving more accurate positioning in communication systems is an urgent issue that needs to be resolved. [Overview of the Initiative] [Problems that the invention aims to solve]

[0003] This application provides a training method for a positioning model, a positioning method, a training device for a positioning model, and a positioning device. The various aspects of this application will be described below. [Means for solving the problem]

[0004] In a first embodiment, a method for training a positioning model is provided, comprising the steps of: obtaining a first channel impulse response (CIR), the first CIR being obtained based on measurement of a reference signal at a first reference point; integrating the first CIR and the first position tag to form first tag data; and training the positioning model based on the first tag data, wherein the reference signal is used for positioning, and the reference signal is transmitted or received by one or more communication devices, including a first communication device.

[0005] In some embodiments, the first location tag includes one or more pieces of information, namely the line-of-sight type between the first reference point and the first communication device, and the location coordinates of the first reference point.

[0006] In some embodiments, the line-of-sight type is determined based on one or more of the distance between the first reference point and the first communication device, and information on the first arrival time at which the reference signal arrives at the first communication device.

[0007] In some embodiments,

Number

Number

Number

Number

Number

Number

[0008] In some embodiments, the first threshold

Number

Number

Number

number

[0009] In some embodiments, the first CIR is a matrix

number

[0010] In some embodiments, the method performs a truncation operation on the first CIR and the matrix

number

[0011] In some embodiments, the method is the matrix

number

number

[0012] In some embodiments, the positioning model includes a convolutional neural network (CNN) model and / or a K-nearest neighbor (KNN) model.

[0013] In the second embodiment, A positioning method is provided, comprising the steps of obtaining a first CIR, the first CIR being obtained based on measurement of a reference signal at a test point, and inputting the first CIR into a positioning model to obtain first position information, wherein the reference signal is used for positioning, and the reference signal is transmitted or received by one or more communication devices, including a first communication device.

[0014] In some embodiments, the first position information includes one or more pieces of information such as the line-of-sight type between the test point and the first communication device, and the similarity probability of the corresponding positions between the test point and the first reference point.

[0015] In some embodiments, the test point is located in a first area, the first area includes a plurality of reference points including the first reference point, and the position coordinates of the test point are determined based on information such as the corresponding similarity probabilities between the test point and the plurality of reference points, and the position coordinates of the plurality of reference points.

[0016] In some embodiments, the position coordinates L of the test point are:

number

[0017] In some embodiments, the first CIR is a matrix

number

[0018] In some embodiments, the method performs a truncation operation on the first CIR and the matrix

number

[0019] In some embodiments, the method is the matrix

number

number

[0020] In some embodiments, the positioning model includes a convolutional neural network (CNN) model and / or a K-nearest neighbor (KNN) model.

[0021] In a third aspect, a training device for a positioning model is provided, comprising a first acquisition unit for acquiring a first channel impulse response (CIR), the first CIR comprising a first acquisition unit obtained based on measurement of a reference signal of a first reference point, an integration unit for integrating the first CIR and the first position tag to form first tag data, and a training unit for training the positioning model based on the first tag data, wherein the reference signal is used for positioning, and the reference signal is transmitted or received by one or more communication devices, including a first communication device.

[0022] In some embodiments, the first location tag includes one or more pieces of information, namely the line-of-sight type between the first reference point and the first communication device, and the location coordinates of the first reference point.

[0023] In some embodiments, the line-of-sight type is determined based on one or more pieces of information, such as the distance between the first reference point and the first communication device, and the first arrival time of the reference signal when it arrives at the first communication device.

[0024] In some embodiments,

number

number

number

number

number

number

[0025] In some embodiments, the first threshold

number

number

number

number

[0026] In some embodiments, the first CIR is a matrix

number

[0027] In some embodiments, the apparatus performs a truncation operation on the first CIR and the matrix

number

[0028] In some embodiments, the apparatus is the matrix

number

number

[0029] In some embodiments, the positioning model includes a convolutional neural network (CNN) model and / or a K-nearest neighbor (KNN) model.

[0030] In a fourth aspect, a positioning device is provided, comprising a second acquisition unit for acquiring a first CIR, the first CIR being obtained based on measurement of a reference signal at a test point, and a processing unit for inputting the first CIR into a positioning model to obtain first position information, wherein the reference signal is used for positioning, and the reference signal is transmitted or received by one or more communication devices, including a first communication device.

[0031] In some embodiments, the first position information includes one or more pieces of information such as the line-of-sight type between the test point and the first communication device, and the similarity probability of the corresponding positions between the test point and the first reference point.

[0032] In some embodiments, the test point is located in a first area, the first area includes a plurality of reference points including the first reference point, and the position coordinates of the test point are determined based on information such as the corresponding similarity probabilities between the test point and the plurality of reference points, and the position coordinates of the plurality of reference points.

[0033] In some embodiments, the position coordinates L of the test point are:

number

[0034] In some embodiments, the first CIR is a matrix

number

[0035] In some embodiments, the apparatus performs a truncation operation on the first CIR and the matrix

number

[0036] In some embodiments, the apparatus is the matrix

number

number

[0037] In some embodiments, the positioning model includes a convolutional neural network (CNN) model and / or a K-nearest neighbor (KNN) model.

[0038] The fifth embodiment provides an apparatus including a memory for storing one or more computer programs and a processor for calling the computer programs in the memory and causing a terminal device to perform some or all of the steps in the first and / or second embodiment.

[0039] The sixth aspect provides an apparatus including a memory for storing one or more computer programs and a processor for calling the computer programs in the memory and causing a network device to perform some or all of the steps in the first and / or second aspect.

[0040] In the seventh aspect, an embodiment of the present application provides a communication system, the system including a device for training the positioning model and / or a device for positioning. In another possible design, the system may further include other devices that interact with the device for training the positioning model or the device for positioning, as in the solution according to the embodiment of the present application.

[0041] In the eighth aspect, an embodiment of the present application provides a computer-readable storage medium storing a computer program that causes a device for training a positioning model and / or a positioning device to perform some or all of the steps in the methods of each of the above aspects.

[0042] In the ninth embodiment, an embodiment of the present application provides a computer program product comprising a non-temporary computer-readable storage medium storing an operable computer program that causes an apparatus for training a positioning model and / or a positioning apparatus to perform some or all of the steps of the methods of each of the above embodiments. In some implementations, the computer program product may be a single software installation package.

[0043] In the tenth embodiment, the embodiment of the present application includes a memory and a processor, the processor being able to call and execute a computer program from the memory, thereby providing a chip that accomplishes some or all of the steps described in the manner of each of the above embodiments. [Effects of the Invention]

[0044] In 5G communication systems, the reference signal (CIR) is easily collected. Therefore, a large amount of CIR data can be acquired. Based on this large amount of CIR data, a more accurate positioning model can be obtained through training. In other words, when positioning is performed using this positioning model, the accuracy of the obtained positioning results is greatly improved. [Brief explanation of the drawing]

[0045] [Figure 1]This is a schematic diagram of the wireless communication system applied in the embodiment of the present invention. [Figure 2] This figure shows an example of a scenario to be applied in the embodiment of the present application. [Figure 3] This is a schematic flowchart of the training method for the positioning model according to the embodiment of the present invention. [Figure 4] This is a schematic flowchart of the positioning method according to the embodiment of the present invention. [Figure 5] This is a schematic flowchart of the method according to Embodiment 1 of the present application. [Figure 6] This is a schematic flowchart of the method according to Embodiment 2 of the present application. [Figure 7] This is a schematic diagram of the training device for a positioning model according to an embodiment of the present invention. [Figure 8] This is a schematic diagram of the positioning device according to an embodiment of the present invention. [Figure 9] This is a schematic diagram of the communication device according to an embodiment of the present invention. [Modes for carrying out the invention]

[0046] The technical solution in this application will be explained below with reference to the drawings.

[0047] Communication system

[0048] Figure 1 shows a wireless communication system 100 applied in an embodiment of the present invention. The wireless communication system 100 may include a network device 110 and a terminal device 120. The network device 110 may be a device that communicates with the terminal device 120. The network device 110 can provide communication coverage to a specific geographic area and can communicate with the terminal device 120 located within that coverage area.

[0049] Figure 1 illustrates one network device and two terminals, and optionally the wireless communication system 100 may include multiple network devices, and the coverage area of ​​each network device may include, but is not limited to, a number of terminal devices in the embodiments of the present application.

[0050] The wireless communication system 100 may optionally further include other network entities such as a network controller or a mobility management entity, but is not limited to these in the embodiments of the present application.

[0051] It can be understood that the technical solutions of the embodiments of this application can be applied to various communication systems, such as 5th generation (5G) systems, new radio (NR), long-term evolution (LTE) systems, LTE frequency division duplex (FDD) systems, and LTE time division duplex (TDD) systems. The technical solutions of this application can also be applied to future communication systems, such as 6th generation mobile communication systems and satellite communication systems.

[0052] The terminal equipment in the embodiments of this application may also be called user equipment (UE), access terminal, user unit, user station, mobile station (MS), mobile terminal (MT), remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent, or user device. The terminal equipment in the embodiments of this application may also refer to a device that provides voice and / or data connectivity to a user, and can be used to connect people, things or devices, such as handheld devices and in-vehicle devices with wireless connectivity. The terminal devices in the embodiments of this application may include mobile phones, tablet PCs (Pads), notebook computers, palmtop computers, mobile internet devices (MIDs), wearable devices, virtual reality (VR) devices, augmented reality (AR) devices, wireless terminals in industrial control, wireless terminals in self-driving, wireless terminals in remote medical surgery, wireless terminals in smart grids, wireless terminals in transportation safety, wireless terminals in smart cities, and wireless terminals in smart homes. Optionally, the UE can be used to function as a base station. For example, the UE can function as a scheduling entity and provide sidelink signals between UEs in vehicle-to-everything (V2X), machine-to-machine (M2M), etc. For example, a cellular phone and a car communicate with each other using sidelink signals. Communication between cellular phones and smart home devices does not require relaying communication signals via base stations.

[0053] In the embodiments of this application, the network equipment may be equipment for communicating with terminal equipment, and the network equipment may include access network equipment, and may also be called access network equipment, wireless access network equipment, or base station, etc. In the embodiments of this application, the access network equipment may refer to a radio access network (RAN) node (or equipment) that provides terminal equipment to a wireless network. Access network equipment broadly covers, or may be replaced by, various names such as NodeB, evolved NodeB (eNB), next-generation NodeB (gNB), relay station, access point, transmitting and receiving point (TRP), transmitting point (TP), master eNB (MeNB), secondary eNB (SeNB), multi-standard radio (MSR) node, home base station, network controller, access node, radio node, access point (AP), transmission node, transceiver node, baseband unit (BBU), remote radio unit (RRU), active antenna unit (AAU), remote radio head (RRH), central unit (CU), distributed unit (DU), and positioning node. A base station may be a macro base station, a micro base station, a relay node, a donor node, or something similar, or a combination thereof. A base station may further refer to a communication module, modem, or chip installed within the aforementioned equipment or device.A base station may also be a mobile switching center and equipment that performs base station functions in D2D, V2X, and machine-to-machine (M2M) communications, network-side equipment in a 6G network, or equipment that performs base station functions in future communication systems. A base station can support networks with the same or different access technologies. The embodiments of this application do not limit the specific technologies and specific equipment forms employed in the access network equipment.

[0054] Base stations may be fixed or mobile. For example, a helicopter or unmanned aerial vehicle may be configured as a mobile base station, with one or more cells moving according to the location of the mobile base station. In another example, a helicopter or unmanned aerial vehicle may be configured as equipment for communicating with another base station.

[0055] In some deployments, the network equipment in the embodiments of the present invention refers to a CU or DU, or the network equipment may include both a CU and a DU. The gNB may further include an AAU.

[0056] Network equipment and terminal equipment may be located on land, including indoors or outdoors, handheld or vehicle-mounted, on water, or in the air on airplanes, balloons, and satellites. The embodiments of this application do not limit the scenarios in which network equipment and terminal equipment are located.

[0057] Communication equipment related to a wireless communication system may include not only access network equipment and terminal equipment, but also core network equipment. Core network equipment may also be network equipment.

[0058] The core network equipment in the embodiments of this application may include equipment that processes and forwards user signaling and data. For example, the core network equipment may include core access and mobility management function (AMF), session management function (SMF), user plane gateway, positioning server, and other core network equipment. The user plane gateway may be a server that has functions such as mobility management, routing, and forwarding of user plane data, and is generally located on the network side, for example, a serving gateway (SGW), packet data network gateway (PGW), or user plane function (UPF). The AMF and SMF may correspond to the mobility management entity (MME) in an LTE system. The AMF is mainly responsible for access, and the SMF is mainly responsible for session management. Naturally, the core network may include other network elements, which will not be listed here.

[0059] The positioning server has a positioning function, and the positioning server according to the embodiments of the present application may include a location management function (LMF) or a location management component (LMC), or it may be a local location management function (LLMF) located in a network device, and the embodiments of the present application are not limited thereto. In some embodiments, the positioning server may be called a location management device.

[0060] It can be understood that all or some of the functions of the communication equipment in this application may be implemented by software functions running on hardware, or by virtualization functions instantiated on a platform (e.g., a cloud platform).

[0061] Positioning technology

[0062] In communication systems, location-based services (such as navigation positioning) are attracting widespread attention. Wireless positioning systems can estimate the location of mobile terminal devices by mapping signal features to spatial locations. Positioning methods are based on one or more of the following: distance measurement, angle, proximity, and fingerprinting. Each of these is explained below.

[0063] In distance-based positioning, a geometric solid (such as a circle or hyperbola) can be constructed based on the distance from the terminal device to at least three network devices. Furthermore, the positioning result can be obtained by calculating the intersection points between these geometric solids. Time of arrival (TOA) and time difference of arrival (TDOA) are signal features commonly used in distance-based positioning methods. Distance-based positioning methods using received signal strength (RSS) require fitting of a path loss model.

[0064] In angle-based positioning, the position of a terminal device can be estimated by calculating the angle of arrival (AOA) of at least two input signals.

[0065] In proximity-based positioning, the location of a terminal device can be determined by the area covered by the network device with the strongest RSS signal received by the terminal device. Because the area covered by network devices is usually wide, this method is simple but has low positioning accuracy.

[0066] Fingerprint-based positioning allows an area to be divided into multiple reference points, and by utilizing environmental features, the location of the reference point with the most similar features can be used as the estimated location of the terminal device. Available signal features may include RSS, etc. Fingerprint-based positioning requires collecting positioning data from each reference point to form a fingerprint. In other words, the accuracy of the fingerprint largely depends on the sufficiency of the dataset. Furthermore, fingerprints have lower hardware requirements compared to distance-based positioning methods. From this, it can be seen that fingerprint-based positioning methods can obtain highly accurate positioning results at the expense of training data collection time.

[0067] Fingerprint positioning algorithms can be divided into probabilistic and deterministic algorithms. Probabilistic algorithms, based on statistical theory such as maximum likelihood estimation (MLE) and / or Bayesian criteria, can estimate the user terminal's position based on the probability density distribution of signal intensity, using the reference point with the highest probability (i.e., the fingerprint). Deterministic algorithms can be divided into two types: those that do not require a training model and those that do. Algorithms that do not require a training model are represented by k-nearest neighbors (KNN) and weighted k-nearest neighbors (WKNN). Algorithms that require a training model can be implemented based on one or more models such as support vector machines (SVM), deep neural networks (DNN), and convolutional neural networks (CNN).

[0068] With the advancement of communication technology, 5G network equipment is being deployed in large numbers, ushering in the 5G-A era. In many environments, terminal devices can easily access the 5G network. Achieving more accurate positioning in communication systems is an urgent issue that needs to be resolved.

[0069] To address the above-mentioned problems, this invention proposes a positioning method. The positioning method according to this invention can obtain a positioning model based on the channel impulse response (CIR) of a reference signal in a communication system. The positioning model can enable the positioning of terminal devices within a target area.

[0070] The positioning model may be an artificial intelligence (AI) model, or a machine learning (ML) model. In other words, this invention can realize AI / ML-based positioning. The model may include, for example, a CNN model, a DNN model, or a KNN model. This invention does not limit the specific structure and parameters of the AI ​​model. Taking a CNN model as an example, the CNN model may include sequentially connected input layers, convolutional layers, pooling layers, fully connected layers, and softmax layers. The parameters of the AI ​​model can be obtained, for example, from the analysis of experimental results.

[0071] The positioning method may include two stages: an offline stage and an online stage.

[0072] The offline phase enables the construction of a positioning model. If the positioning model is an AI model, the construction of the positioning model can be achieved through learning or training. Therefore, the offline phase may also be called the training phase or learning phase.

[0073] In the offline phase, a positioning model can be constructed using a first CIR. The first CIR may be obtained based on measurements of a reference signal (RS). The first CIR may be obtained by measuring the reference signal transmitted at a first reference point. That is, the first CIR may refer to a measured value of the CIR of the reference signal transmitted or received at the first reference point. Correspondingly, the CIRs of multiple reference points may be obtained by measuring the reference signals transmitted or received at each of the multiple reference points.

[0074] The first reference point may belong to multiple reference points. These multiple reference points may be acquired from the positioning target area. This application does not restrict the method of selecting multiple reference points. For example, reference points may be uniformly selected within the positioning target area.

[0075] The positioning model may be used to determine the location information of a terminal device. Therefore, the positioning model can be trained (built) based on the first CIR and the location information of the first reference point, using the location information of the first reference point (i.e., the actual location information of the first reference point) as a tag. It is understood that the more training data there is for training the positioning model, the higher the accuracy of the positioning model will be. Therefore, the training data may include other data as well; that is, the training data may include not only the first CIR and the location information of the first reference point, but also the CIRs of other reference points and the location information of other reference points. For example, the location information of some or all of the reference points among multiple reference points can be organized into a fingerprint library, and the positioning model can be trained using the fingerprint library. Some or all of the reference points among multiple reference points may include the first reference point. In some embodiments, a relatively ideal positioning model can be obtained through simulation.

[0076] During the online phase, test data can be used to determine the location of a test point where a terminal device is located. The test point may be a single location within the positioning area. That is, if the test point where the terminal device is located is within the positioning area, the positioning model can determine the location information of the test point. The test data may include, for example, the CIR of the test point. That is, at the test point, the CIR of the test point can be measured, and the location information of the test point can be obtained based on the CIR.

[0077] It can be understood that the online phase may also be the phase in which the positioning model is used. Therefore, the online phase may also be called the positioning phase.

[0078] The positioning target area may be an area to which the above positioning model can be applied, or an area where the positioning model is effective. The positioning target area may be located indoors.

[0079] In some communication systems (such as 5G systems), the CIR of the reference signal is easily collected. Therefore, a large amount of CIR data can be acquired. Based on this large amount of CIR data, the positioning model constructed can be made more accurate. The positioning accuracy achieved based on this positioning model is significantly improved.

[0080] The following describes the input and output data for the positioning model.

[0081] The input to the positioning model may be a first CIR. In the offline phase, the first CIR may be obtained based on measurements of a reference signal at a first reference point. For example, at some or all of the reference points in the positioning area, terminal equipment can be used to collect CIR measurements of the reference signal at each point. The first CIR may be any one of the multiple collected CIR measurements. In the online phase, the first CIR may be obtained based on measurements of a reference signal at a test point. For example, in the positioning area, test points can be randomly selected, and terminal equipment can be used to collect the first CIR of the reference signal.

[0082] In this application, the reference signal may be a reference signal used for positioning. For example, the reference signal may be one or more of the sounding reference signal (SRS), positioning reference signal (PRS), etc., used for positioning.

[0083] A reference signal (or reference signal sequence) may be transmitted or received by one or more communication devices. One or more communication devices may include a first communication device. The first communication device may be used for positioning a terminal device to be positioned. For example, the first communication device may include a base station that transmits a reference signal. In one embodiment, the base station can transmit a reference signal, and the terminal device can receive a reference signal. The terminal device can measure the reference signal to obtain a first CIR. The terminal device can send the measured first CIR back to the positioning server. The positioning server can perform calculations based on the first CIR using a positioning model.

[0084] Furthermore, this application does not limit the type of the first communication device. For example, the first communication device may be a base station that receives a reference signal, or it may be another terminal device that performs side-link communication with the terminal device to be positioned.

[0085] This application does not restrict the method of representing the first CIR. The following explanation will use the representation of the first CIR as a matrix as an example.

[0086] In some embodiments, the first CIR is a matrix

number

number

number

[0087] It can be understood that a multipath effect exists between the transmitting and receiving ends of the reference signal. Therefore, the receiving end can perform multiple samplings to comprehensively represent the first CIR. The number of sampling points corresponding to multiple samplings may be N as described above. N may be a positive integer. For example, N may be 4,096.

[0088] In some embodiments, the first CIR may be preprocessed. Preprocessing can be used to obtain a processed first CIR. Inputting the processed first CIR into the positioning model simplifies the processing of the positioning model, thereby improving the processing efficiency of the positioning model.

[0089] In some embodiments, the preprocessing may include, for example, truncation. The truncation may include reducing the number of sampling points in the first CIR. For example, the number of sampling points in the first CIR can be reduced by deleting sampling points that are close to zero or by deleting sampling points that have little impact on the data. The truncation can reduce the amount of data in the first CIR, thereby simplifying the computational complexity of the positioning model and improving the processing efficiency of the positioning model.

[0090] The first CIR is a matrix

number

number

number

number

number

number

number

[0091] In some embodiments, the preprocessing may include row-to-column transformations. For example, matrix

number

number

number

number

[0092] Furthermore, if there are missing elements in the matrix after the row-to-column conversion, it can be filled with zero elements. For example,

number

number

[0093] Furthermore, as mentioned above

number

number

[0094] The input of the positioning model has been described in detail above, and the output of the positioning model will be described below.

[0095] The output of the positioning model may be the first position information. The first position information may correspond to the first CIR. The position point corresponding to the first position information may be the first reference point or the test point. That is, when the first CIR is the CIR of the first reference point, the first position information may be the position information of the first reference point, and when the first CIR is the CIR of the test point, the first position information may be the position information of the test point. In the online stage, after the first CIR is input into the positioning model, the positioning model can derive and output the first position information. In the offline stage, the training data for training the positioning model may include the first position information, that is, the first position information can be used to train the positioning model as a tag. Therefore, in the offline stage, the first position information may be referred to as the first position tag. Hereinafter, the first position information will be mainly exemplified, and the first position tag can refer to the description of the first position information.

[0096] The first position information may include any information related to the position. In some embodiments, the first position information may include one or more of the line-of-sight type between the position point corresponding to the first position information and the first communication device, and the position coordinates of the position point corresponding to the first position information.

[0097] The positioning target area can be represented by an area in a coordinate system. In this case, the position coordinates can be represented by coordinate points in the coordinate system. For example, the position coordinates may be represented as (x b , y b ). x b may represent the x-axis coordinate corresponding to the position coordinates, and y b may represent the y-axis coordinate corresponding to the position coordinates. When the first position information includes the position coordinates, it can be understood that the positioning model may be used to determine the specific coordinates of the terminal device, that is, the positioning model may be directly used for positioning.

[0098] In one embodiment, the positioning model can output the similarity between a test point and a first reference point, and the position coordinates of the test point can be determined by the similarity. That is, the present invention transforms the positioning problem into a problem of similarity between the position of a test point and each reference point. When the positioning target area includes multiple reference points, the positioning model can output the similarity between the test point and the multiple reference points. The similarity can be expressed as a similarity probability. The similarity probability of corresponding positions between the test point and the first reference point may be used to represent the probability that the test point is located at the position where the first reference point is located. For example, in K reference points, the similarity between the test point and the k-th reference point is probability p k It may also be expressed as follows.

[0099] In some examples, highly similar location points can be grouped into the same category. For example, the positioning coordinates L of a test point can be obtained using a probabilistic weighted centroid algorithm. The probabilistic weighted centroid algorithm is expressed by the formula:

number

[0100] In the probability-weighted centroid algorithm, the K reference points may be some or all of the reference points in the positioning target area. For example, the K reference points may be some of the reference points with high similarity to the test point. In one embodiment, the reference points output by the positioning model can be sorted according to their similarity probability, and the first K reference points with the largest similarity probability can be selected as the reference points to participate in the probability-weighted centroid algorithm.

[0101] If the position coordinates of a reference point are used directly as the position coordinates of a test point, it can be understood that the position coordinates of the test point are limited to the position coordinates of a specific reference point. However, the coordinates of the test point obtained by the probabilistic weighted centroid algorithm may be any coordinate point within the positioning target area. Therefore, based on the probabilistic weighted centroid algorithm proposed in this application, a more accurate position can be obtained, thereby improving positioning accuracy.

[0102] Line-of-sight types can be divided into non-line of sight (NLOS) and line of sight (LOS). Line of sight (LOS) may refer to the terminal device being positioned and the first communication device being visible to each other or via visual aids. Non-line of sight (NLOS) may refer to the terminal device being positioned and the first communication device being invisible to each other or via visual aids. Figure 2 shows examples of line of sight (LOS) and non-line of sight (NLOS). As shown in Figure 2, terminal device 121 is the terminal device being positioned. Terminal device 121 is located in room 1. Both network devices 111 and 112 are located in room 1. Since they are located in the same room and network device 111 or network device 112 and terminal device 121 are visible to each other, the line-of-sight type between network device 111 and terminal device 121 can be classified as line of sight (LOS), and the line-of-sight type between network device 112 and terminal device 121 can also be classified as line of sight (LOS). The network device 113 is located in room 2. Because room 1 and room 2 are separated by a wall, the network device 113 and the terminal device 121 are not easily visible to each other. In this case, the line-of-sight type between the network device 113 and the terminal device 121 can be classified as non-line-of-sight (NLOS).

[0103] In some embodiments, the positioning location cannot be accurately calculated because the line-of-sight type cannot be determined. For example, in the case of non-line-of-sight (NLOS), the reference signal may be shielded or reflected, so positioning based on the corresponding reference signal may actually decrease in accuracy. Continuing to refer to Figure 2, the position of terminal device 121 can be jointly determined by reference signals transmitted by network devices 111, 112, and 113. However, because non-line-of-sight (NLOS) conditions exist, positioning in combination with network device 113 may significantly decrease in accuracy.

[0104] According to this invention, the line-of-sight type can be determined, and an appropriate positioning algorithm or data can be selected based on the line-of-sight type. For example, non-line-of-sight (NLOS) type data can be excluded from positioning data. Referring again to Figure 2, according to this invention, if it is determined that there is a non-line-of-sight (NLOS) condition between the network device 113 and the terminal device 121, the reference signal transmitted by the network device 113 is ignored, and the position of the terminal device can be determined based only on the reference signals of the network devices 111 and 112. This avoids the adverse effects of non-line-of-sight (NLOS) conditions on positioning, thereby improving positioning accuracy.

[0105] As will be clear from this, when the first location information includes a line-of-sight type, the positioning model may be used to determine whether the connection between the terminal device and the first network device is line-of-sight (LOS) or non-line-of-sight (NLOS). Furthermore, an appropriate positioning algorithm or data can be selected based on the line-of-sight type. That is, data or positioning algorithm screening can be performed based on the line-of-sight type to achieve accurate positioning. In other words, in this case, the positioning model may be used indirectly for positioning.

[0106] As described above, after determining the line-of-sight type, the position coordinates of the test points can be further determined based on the line-of-sight type. This application does not limit the method for determining the position coordinates of the test points. For example, the position coordinates of the test points can be determined by the direct positioning method described above. Alternatively, the position coordinates of the test points may be determined by positioning algorithms such as TOA, TDOA, or path loss.

[0107] This application does not limit the representation methods for line-of-sight types. For example, line-of-sight types are:

number

number

number

[0108] As described above, in the online phase, the line-of-sight type of the test point can be determined based on the positioning model. In the offline phase, the present invention provides a method for determining the line-of-sight type, and the determined line-of-sight type can be used as a tag for training data to further train the positioning model.

[0109] In one embodiment, the line-of-sight type may be associated with one or more pieces of information, such as the distance between a first reference point and a first communication device, and the first arrival time of the reference signal when it arrives at the first communication device.

[0110] The initial arrival time may be the length of time from the time the transmitter sends the reference signal to the time the receiver first receives the reference signal. For example, the initial arrival time may be:

number

number

[0111] In some embodiments,

number

number

number

number

[0112] Furthermore, both the first and second thresholds can be set flexibly. For example, both the first and second thresholds may be small values ​​close to 0. In one embodiment, either the first or second threshold is c × t s It's fine. s This may be the sampling interval of the reference signal. Also, the first threshold and the second threshold may be the same or different. When the first threshold and the second threshold are the same, the line-of-sight type determination method is:

number

[0113] To make it easier to understand, the line-of-sight type can be line of sight (LOS) if the difference between the product of the initial arrival time and the speed of light and the distance from the first reference point to the first communication device is less than the first threshold. In other words, the line-of-sight type can be line of sight (LOS) if the difference between the product of the initial arrival time and the speed of light and the distance from the first reference point to the first communication device is relatively small. The line-of-sight type can be out-of-line-of-sight (NLOS) if the difference between the product of the initial arrival time and the speed of light and the distance from the first reference point to the first communication device is greater than the second threshold. In other words, the line-of-sight type can be out-of-line-of-sight (NLOS) if the difference between the product of the initial arrival time and the speed of light and the distance from the first reference point to the first communication device is relatively large.

[0114] In the offline phase, the first CIR can be integrated with the first location tag to form the first tag data. For example, if the first location tag includes the location coordinates where the first reference point is located, the first tag data will be:

number

number

[0115] In the case of multiple reference points, the tag data corresponding to each reference point can constitute a tag dataset. For example, if the first location tag includes location coordinates, the tag dataset may be represented as F=(Txy). For example, if the first location tag includes line-of-sight type, the tag dataset may be represented as F=(TG). Since the tag dataset is determined based on the reference points, the tag dataset may be referred to as a fingerprint library. Correspondingly, the positioning model may be implemented based on fingerprints.

[0116] The steps in the offline and online phases will be explained below with reference to Figures 3 and 4.

[0117] Figure 3 is a schematic flowchart of a training method for a positioning model according to an embodiment of the present invention. The method shown in Figure 3 can be performed by electronic equipment. The electronic equipment may be, for example, the communication equipment described above. For example, the method shown in Figure 3 can be performed by a positioning server. The method shown in Figure 3 may include steps S310 to S330.

[0118] In step S310, the first CIR is obtained.

[0119] In step S320, the first CIR and the first location tag are integrated to form the first tag data.

[0120] In step S330, the positioning model is trained based on the first tag data.

[0121] This application does not limit the specific training methods for the positioning model. For example, the positioning model may be trained using methods such as genetic algorithms or neural evolution algorithms.

[0122] Figure 4 is a schematic flowchart of a positioning method according to an embodiment of the present invention. That is, Figure 4 is a schematic diagram of the online stage. The method shown in Figure 4 can be performed by electronic equipment. The electronic equipment may be, for example, the communication equipment described above. For example, the method shown in Figure 4 can be performed by a positioning server. The method shown in Figure 4 may include steps S410 to S420.

[0123] In step S410, the first CIR is obtained.

[0124] In step S420, the first CIR is input to the positioning model to obtain the first position information.

[0125] To facilitate understanding of this application, the application will be described in detail below through Examples 1 and 2.

[0126] Example 1

[0127] Figure 5 shows an example of a direct positioning method according to Embodiment 1. Based on the method shown in Figure 5, direct positioning of a terminal device can be achieved, that is, the position coordinates of the terminal device can be output.

[0128] The method shown in Figure 5 may include an offline phase and an online phase. In the offline phase, data collection and model training can be performed. In the online phase, data collection and terminal device positioning can be performed.

[0129] The offline phase may include steps S511 to S516.

[0130] In step S511, there are A base stations in the positioning target area, B reference points are randomly selected in the area, a mobile terminal is placed at each reference point, and the CIR measurement matrix of each base station is determined.

number

[0131] CIR measurement values ​​are

number

number

number

[0132] In step S512, the CIR measurement matrix for each base station is displayed.

number

number

number

number

[0133] In step S513, the truncated measurement matrix is ​​used.

number

number

number

number

[0134] In other words, in step S513, a matrix of size M×2 can be transformed into a matrix of size 2u×v+1, and missing elements of 0 can be filled in. The values ​​of u and v can be obtained by experimental comparison and parameter tuning. For example, u=256 and uv+1 is the number of points in the dataset.

[0135] In step S514, the position coordinates (x) of each corresponding reference point are b ,y b ) can be obtained.

[0136] In step S515, the CIR images obtained at each reference point

Number

Number

[0137] In step S516, the position fingerprint library F is used as the input of the CNN model, and the CNN model is trained. After training the CNN model, a CNN position classification model is obtained.

[0138] The CNN model may include an input layer, a convolutional layer, a pooling layer, a fully connected layer, and a softmax layer connected in sequence. The parameters of the CNN model can be obtained from the analysis of experimental results.

[0139] The online stage may include steps S521 to S524.

[0140] In step S521, test points are selected in the positioning target area, and the mobile terminal measures the CIR measurement matrix of each base station at the test points

Number

[0141] In step S522, the CIR measurement matrix of each base station

Number

Number

Number

[0142] In step S523, perform row and column transformation on the measured value matrix

Number

Number

[0143] The process of obtaining the CIR image in steps S521 to S523 is the same as the process in steps S511 to S513, and will not be repeatedly described here.

[0144] In step S524, input the CIR image

Number

[0145] The weighted centroid algorithm can be realized by the formula

Number

[0146] Ω is the set of the first K reference points with high similarity probability, (x k , y k ) is the coordinate of the k-th reference point. k = 1, 2,..., K.

[0147] Example 2

[0148] Figure 6 shows an example of an indirect positioning method according to Embodiment 2. Based on the method shown in Figure 6, indirect positioning of a terminal device can be achieved, that is, the line-of-sight type of the terminal device can be output. In other words, the method according to Figure 6 may also be a method for classifying line-of-sight types.

[0149] The method shown in Figure 6 may include an offline phase and an online phase. In the offline phase, data collection and model training can be performed. In the online phase, data collection and terminal device positioning can be performed.

[0150] The offline phase may include steps S611 to S614.

[0151] In step S611, there are A base stations in the positioning target area, B reference points are randomly selected in the area, mobile terminals are placed at each reference point, and the CIR measurement matrix of each base station's reference point is determined.

number

[0152] CIR measurement is matrix

number

number

number

[0153] Step S612 determines the line-of-sight (LOS) / non-line-of-sight (NLOS) conditions between the reference point and the base station.

number

[0154] The Line of Sight (LOS) / Non-Line of Sight (NLOS) tags are:

number

[0155] CIR measurement matrix collected at each base station

number

number

number

[0156] c is the speed of light,

number

number

number

[0157] In step S613, the CIR measurement matrix obtained at each reference point is obtained.

number

number

number

[0158] In step S614, the tag dataset F is used as input to the line of sight type classification model (hereinafter abbreviated as LOS classification model), and the LOS classification model is obtained by training.

[0159] The online phase may include steps S621 and S622. In step S621, a test point is selected within the positioning area, and the mobile terminal is located at the test point and displays the CIR measurement matrix of each base station.

number

[0160] Step S622 is the CIR measurement matrix of each base station collected at the test point.

number

[0161] The embodiments of the method of this application are described in detail above, and embodiments of the apparatus of this application are described in detail below. Since the description of the embodiment of the method corresponds to the description of the embodiment of the apparatus, please understand that you can refer to the embodiment of the method above for parts that are not described in detail.

[0162] Figure 7 is a schematic diagram of the training device 700 for a positioning model according to an embodiment of the present invention. The device 700 includes a first acquisition unit 710, an integration unit 720, and a training unit 730.

[0163] The first acquisition unit 710 may be used to acquire a first CIR, which is obtained based on the measurement of a reference signal at a first reference point.

[0164] The integration unit 720 may be used to integrate the first CIR and the first location tag to form the first tag data.

[0165] The training unit 730 may be used to train a positioning model based on the first tag data.

[0166] The reference signal is used for positioning, and is transmitted or received by one or more communication devices, including the first communication device.

[0167] In some embodiments, the first location tag includes one or more pieces of information, such as the line-of-sight type between the first reference point and the first communication device, and the location coordinates of the first reference point.

[0168] In some embodiments, the line-of-sight type is determined based on one or more pieces of information, such as the distance between a first reference point and a first communication device, and the first arrival time of the reference signal when it arrives at the first communication device.

[0169] In some embodiments,

number

number

number

number

number

number

[0170] In some embodiments, the first threshold

number

number

number

number

[0171] In some embodiments, the first CIR is a matrix

number

[0172] In some embodiments, the device performs a truncation operation on the first CIR and the matrix

number

[0173] In some embodiments, the apparatus is a matrix

number

number

[0174] In some embodiments, the positioning model includes a CNN model and / or a KNN model.

[0175] Figure 8 shows a positioning device 800 according to an embodiment of the present application. In one embodiment, the device 800 may be a communication device. The device 800 includes a second acquisition unit 810 and a processing unit 820.

[0176] The second acquisition unit 810 may be used to acquire the first CIR, which is obtained based on the measurement of a reference signal at a test point.

[0177] The processing unit 820 may be used to input the first CIR into a positioning model to obtain first position information.

[0178] The reference signal is used for positioning, and is transmitted or received by one or more communication devices, including the first communication device.

[0179] In some embodiments, the first position information includes one or more pieces of information such as the line-of-sight type between the test point and the first communication device, and the similarity probability of the corresponding positions between the test point and the first reference point.

[0180] In some embodiments, the test point is located in a first area, which includes a plurality of reference points, including a first reference point, and the position coordinates of the test point are determined based on information such as the corresponding similarity probabilities between the test point and the plurality of reference points, and the position coordinates of the plurality of reference points.

[0181] In some embodiments, the position coordinates L of the test point are:

number

[0182] In some embodiments, the first CIR is a matrix

number

[0183] In some embodiments, the device performs a truncation operation on the first CIR and the matrix

number

[0184] In some embodiments, the apparatus is a matrix

number

number

[0185] In some embodiments, the positioning model includes a CNN model and / or a KNN model.

[0186] In an optional embodiment, the first acquisition unit 710, the integration unit 720, the training unit 730, the second acquisition unit 810, or the processing unit 820 may be a processor 910. Specifically, as shown in Figure 9, the device 700 or device 800 may further include a transceiver 930 and / or memory 920.

[0187] Figure 9 is a schematic diagram of a communication device in an embodiment of the present application. The dashed lines in Figure 9 indicate that the unit or module is selectable. The device 900 can be used to implement the method described in the above embodiment. The device 900 may be an electronic device, a chip, a terminal device, or a network device.

[0188] The apparatus 900 may include one or more processors 910. The processors 910 can support the apparatus 900 in implementing the methods described in the above embodiment of the method. The processors 910 may be general-purpose processors or dedicated processors. For example, the processor may be a central processing unit (CPU). Alternatively, the processor may be another general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, etc. The general-purpose processor may be a microprocessor, or the processor may be any conventional processor, etc.

[0189] The device 900 may further include one or more memories 920. A program is stored in the memory 920, and the program is executable by the processor 910, causing the processor 910 to execute the method described in the above embodiment of the method. The memory 920 may be independent of the processor 910 or may be integrated with the processor 910.

[0190] The device 900 may further include a transceiver 930. The processor 910 can communicate with other devices or chips via the transceiver 930. For example, the processor 910 can send and receive data with other devices or chips via the transceiver 930.

[0191] Embodiments of the present application further provide a computer-readable storage medium for storing a program. The computer-readable storage medium can be applied to the apparatus according to the embodiments of the present application, and the program causes a computer to execute the method performed by the apparatus in each embodiment of the present application.

[0192] Embodiments of the present application further provide a computer program product, which includes a program, which can be applied to an apparatus according to an embodiment of the present application, and which causes a computer to execute a method performed by the apparatus in each embodiment of the present application.

[0193] Embodiments of the present application further provide a computer program. The computer program can be applied to the apparatus according to the embodiments of the present application, and the computer program causes a computer to execute the methods performed by the apparatus in each embodiment of the present application.

[0194] It is understandable that, in this application, the terms “system” and “network” may be interchangeable. Furthermore, the terms used in this application are used solely to interpret the specific embodiments of this application and are not intended to limit it. The terms “first,” “second,” “third,” and “fourth,” etc., in the specification, claims, and drawings of this application are used to distinguish different subjects, not to describe a specific order. Also, the terms “include,” “have,” and any variations thereof are intended to cover non-exclusive inclusion.

[0195] In the embodiments of the present application, the “instruction” referred to may be a direct instruction, an indirect instruction, or an indication of a related relationship. For example, A instructing B may mean that A directly instructs B, for example, indicating that B can be obtained by A; or A indirectly instructs B, for example, that A instructs C, indicating that B can be obtained by C; or an indication of a related relationship between A and B.

[0196] In the embodiments of this application, "B corresponding to A" indicates that B is associated with A and that B can be determined in accordance with A. However, determining B in accordance with A does not mean determining B in accordance with A alone, but rather that B may be determined in accordance with A and / or other information.

[0197] In the embodiments of this application, the term "correspondence" may indicate a direct or indirect correspondence between the two, a related relationship between the two, or a relationship such as instruction and instruction, setting and setting.

[0198] In the embodiments of this application, “pre-defined” or “pre-configured” may be implemented by pre-storing in a device (including, for example, terminal devices and network devices) a form that can indicate the corresponding code, form, or related information, and this application does not limit the specific form of such implementation. For example, pre-defined may refer to something defined in a protocol.

[0199] In the embodiments of this application, “protocol” may refer to a standard protocol in the field of communications, and may include, for example, the LTE protocol, the NR protocol, and related protocols applicable to future communications systems, but is not limited to this.

[0200] In the embodiments of this application, the term "and / or" simply describes the relationship between related objects and indicates that three types of relationships exist. For example, A and / or B include the three situations where only A exists, where A and B exist simultaneously, and where only B exists. In this specification, the symbol " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0201] In the embodiments of this application, “includes” may refer to direct or indirect implication. Optionally, “includes” as used in the embodiments of this application may be replaced with “indicates” or “used to determine.” For example, “A includes B” may be replaced with “A indicates B” or “used to determine B.”

[0202] In the various embodiments of the present application, the magnitude of the process numbers does not indicate the order of execution, and the execution order of each process should be determined based on its function and inherent logic, and does not constitute any limitation on the implementation processes of the embodiments of the present application.

[0203] In some embodiments relating to this application, it should be understood that the disclosed systems, devices, and methods can be implemented in other forms. For example, the device embodiments described above are merely schematic, and for instance, the division of units is merely one type of logic function division. In actual implementations, other division methods may be adopted, for example, multiple units or components may be combined or integrated into another system, or some features may be ignored or omitted. Furthermore, the mutual coupling, direct coupling, or communication connection shown or considered may also be an indirect coupling or communication connection via some interface, device, or unit, and may be in the form of electrical, mechanical, or other means.

[0204] The units described as separating members may or may not be physically separated, and the members shown as units may or may not be physical units; that is, they may be located in one place or distributed among multiple network units. Some or all of the units can be selected as needed to achieve the objectives of the means of this embodiment.

[0205] Furthermore, each functional unit in each embodiment of the present application may be integrated into a single processing unit, each unit may exist physically separately, and two or more units may be integrated into a single unit.

[0206] In the embodiments described above, all or part of the embodiments may be implemented by software, hardware, firmware, or any combination thereof. If implemented by software, all or part of the embodiments may be implemented in the form of a computer program product. A computer program product includes one or more computer instructions. Loading and executing the computer program instructions on a computer generates all or part of the procedures or functions described in the embodiments of this application. The computer may be a general-purpose computer, a dedicated computer, a computer network, or other programmable device. The computer instructions may be stored on a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wire (e.g., coaxial cable, fiber optic cable, digital subscriber line (DSL)) or wirelessly (e.g., infrared, radio, microwave, etc.). The computer-readable storage medium may be any available medium that a computer can read, or a data storage device such as a server or data center that integrates one or more available media. The available media may be magnetic media (e.g., floppy disks, hard disks, tapes), optical media (e.g., digital video discs (DVDs)), or semiconductor media (e.g., solid state disks (SSDs)).

[0207] Although specific embodiments of the present application have been described above, the scope of protection of this application is not limited thereto. Any modifications or substitutions that a person skilled in the art could easily conceive without departing from the technical scope disclosed herein fall within the scope of protection of this application. Therefore, the scope of protection of this application should be the same as the scope of protection of the claims. [Explanation of Symbols]

[0208] 100 Wireless Communication Systems 110 Network Equipment 111 Network Equipment 112 Network Equipment 113 Network Equipment 120 terminal devices 121 Terminal equipment 700 Training Devices 710 First Acquisition Unit 720 Integrated Units 730 Training Units 800 equipment 810 Second Acquisition Unit 820 processing units 900 equipment 910 Processor 920 memory 930 Transmitter / Receiver

Claims

1. A training method for positioning models, A step of acquiring a first channel impulse response (CIR), wherein the first CIR is obtained based on a measurement of a reference signal at a first reference point. The steps include: integrating the first CIR and the first position tag to form first tag data; A step of training the positioning model based on the first tag data, Includes, A method for training a positioning model, characterized in that the reference signal is used for positioning, and the reference signal is transmitted or received by one or more communication devices, including a first communication device.

2. The first position tag is, The line-of-sight type between the first reference point and the first communication device, and One or more pieces of information regarding the position coordinates of the first reference point. The method according to claim 1, characterized by including the following:

3. The aforementioned forecast type is, The distance between the first reference point and the first communication device, and The reference signal is determined based on one or more pieces of information, such as the first arrival time of the reference signal when it arrives at the first communication device. The method according to feature 2. 【Request Item 4】 【Number 128】 If the following conditions are met, the line-of-sight type is line of sight (LOS) and / or [Number 129] If the conditions are met, the line-of-sight type is outside of line of sight (NLOS), c is the speed of light. [Number 130] The aforementioned first arrival time, [Number 131] is the first threshold, [Number 132] This is the second threshold, [Number 133] The method according to the previous claim, characterized in that b is the distance from the first reference point to the first communication device, b is the identifier of the first reference point, and a is the identifier of the first communication device.

5. The first threshold [Number 134] teeth, [Number 135] Satisfying the conditions and / or the second threshold [Number 136] teeth, [Number 137] Satisfying t ｓ The method according to 4, characterized in that is the sampling interval of the reference signal.

6. The first CIR is a matrix. [Number 138] The method according to any one of claims 1 to 5, characterized in that it is expressed as such, where N is the number of sampling points corresponding to the first CIR, R is the real part, I is the imaginary part, b is the identifier of the first reference point, and a is the identifier of the first communication device.

7. The first CIR is truncated, and the matrix is ​​formed. [Number 139] The process further includes the step of obtaining a processed first CIR represented by The method according to 6, characterized in that M is a value less than N.

8. The aforementioned matrix [Number 140] The process further includes the step of performing row and column transformations to obtain a CIR image, The aforementioned CIR image is a matrix. [Number 141] The method according to 7, characterized in that it is represented as such, where u is the number of sampling points corresponding to the processed first CIR, and 2u × v + 1 represents the number of points in the dataset.

9. The method according to any one of claims 1 to 8, characterized in that the positioning model includes a convolutional neural network (CNN) model and / or a K-nearest neighbor (KNN) model.

10. A method for positioning, A step of acquiring a first channel impulse response (CIR), wherein the first CIR is obtained based on a measurement of a reference signal at a test point. The first CIR is input to the positioning model to obtain first position information, Includes, A positioning method characterized in that the reference signal is used for positioning, and the reference signal is transmitted or received by one or more communication devices, including a first communication device.

11. The first position information is, The line-of-sight type between the test point and the first communication device, and One or more pieces of information regarding the similarity probability of the corresponding positions between the aforementioned test point and the first reference point. The method according to the present invention, characterized by including the following:

12. The method according to 11, characterized in that the test point is located in a first area, the first area includes a plurality of reference points including the first reference point, and the position coordinates of the test point are determined based on information such as the corresponding similarity probabilities between the test point and the plurality of reference points, and the position coordinates of the plurality of reference points.

13. The position coordinates L of the aforementioned test point are: [Number 142] The condition satisfies, and Ω is the first set of K reference points with high similarity probability, (x ｋ , y ｋ ) is the coordinate of the k-th reference point, p ｋ The method according to 12, characterized in that is the similarity probability of the corresponding positions of the test point and the k-th reference point.

14. The first CIR is a matrix. [Number 143] The method according to any one of claims 10 to 13, characterized in that it is expressed as such, where N is the number of sampling points corresponding to the first CIR, R is the real part, I is the imaginary part, b is the identifier of the test point, and a is the identifier of the first communication device.

15. The first CIR is truncated, and the matrix is ​​formed. [Number 144] The process further includes the step of obtaining a processed first CIR represented by The method according to 14, characterized in that M is a value less than N.

16. The aforementioned matrix [Number 145] The process further includes the step of performing row and column transformations to obtain a CIR image, The aforementioned CIR image is a matrix. [Number 146] The method according to 15, characterized in that it is represented as such, where u is the number of sampling points corresponding to the processed first CIR, and 2u × v + 1 represents the number of points in the dataset.

17. The method according to any one of claims 10 to 16, characterized in that the positioning model includes a convolutional neural network (CNN) model and / or a K-nearest neighbor (KNN) model.

18. A training device for positioning models, A first acquisition unit for acquiring a first channel impulse response (CIR), wherein the first CIR is obtained based on the measurement of a reference signal at a first reference point, and the first acquisition unit An integration unit for integrating the first CIR and the first position tag to form first tag data, A training unit for training the positioning model based on the first tag data, Includes, A training device for a positioning model, characterized in that the reference signal is used for positioning, and the reference signal is transmitted or received by one or more communication devices, including a first communication device.

19. The first position tag is, The line-of-sight type between the first reference point and the first communication device, and One or more pieces of information regarding the position coordinates of the first reference point. The apparatus according to claim 18, characterized by including the following:

20. The aforementioned forecast type is, The distance between the first reference point and the first communication device, and One or more pieces of information regarding the first arrival time of the reference signal when it arrives at the first communication device. The apparatus according to claim 19, characterized in that it is determined based on the following. [Request Item 21] [Number 147] If the following conditions are met, the line-of-sight type is line of sight (LOS) and / or [Number 148] If the conditions are met, the line-of-sight type is outside of line of sight (NLOS), c is the speed of light. [Number 149] The aforementioned first arrival time, [Number 150] is the first threshold, [Number 151] This is the second threshold, [Number 152] The apparatus according to claim 20, characterized in that b is the distance from the first reference point to the first communication device, b is the identifier of the first reference point, and a is the identifier of the first communication device.

22. The first threshold [Number 153] teeth, [Number 154] Satisfying the conditions and / or the second threshold [Number 155] teeth, [Number 156] Satisfying t ｓ The apparatus according to claim 21, characterized in that is the sampling interval of the reference signal.

23. The first CIR is a matrix. [Number 157] The apparatus according to any one of claims 18 to 22, characterized in that it is represented as such, where N is the number of sampling points corresponding to the first CIR, R is the real part, I is the imaginary part, b is the identifier of the first reference point, and a is the identifier of the first communication device.

24. The first CIR is truncated, and the matrix is ​​formed. [Number 158] It further includes a first truncation unit for obtaining a processed first CIR represented by The apparatus according to claim 23, characterized in that M is a value less than N.

25. The aforementioned matrix [Number 159] The system further includes a first transformation unit for performing row and column transformations to obtain a CIR image, The aforementioned CIR image is a matrix. [Number 160] The apparatus according to claim 24, characterized in that it is represented as such, where u is the number of sampling points corresponding to the processed first CIR, and 2u × v + 1 represents the number of points in the dataset.

26. The apparatus according to any one of claims 18 to 25, characterized in that the positioning model includes a convolutional neural network (CNN) model and / or a K-nearest neighbor (KNN) model.

27. A positioning device, A second acquisition unit for acquiring a first CIR, wherein the first CIR is obtained based on the measurement of a reference signal at a test point, and the second acquisition unit A processing unit for inputting the first CIR into a positioning model to obtain first position information, Includes, A positioning device characterized in that the reference signal is used for positioning, and the reference signal is transmitted or received by one or more communication devices, including a first communication device.

28. The first position information is, The line-of-sight type between the test point and the first communication device, and One or more pieces of information regarding the similarity probability of the corresponding positions between the aforementioned test point and the first reference point. The apparatus according to claim 27, characterized by including the following:

29. The apparatus according to claim 28, characterized in that the test point is located in a first area, the first area includes a plurality of reference points including the first reference point, and the position coordinates of the test point are determined based on information such as the corresponding similarity probability between the test point and the plurality of reference points, and the position coordinates of the plurality of reference points.

30. The position coordinates L of the aforementioned test point are: [Number 161] The condition satisfies, and Ω is the first set of K reference points with high similarity probability, (x ｋ , y ｋ ) is the coordinate of the k-th reference point, p ｋ The apparatus according to claim 29, wherein is the similarity probability of the corresponding positions of the test point and the k-th reference point.

31. The first CIR is a matrix. [Number 162] The apparatus according to any one of claims 27 to 30, characterized in that it is represented as such, where N is the number of sampling points corresponding to the first CIR, R is the real part, I is the imaginary part, b is the identifier of the test point, and a is the identifier of the first communication device.

32. The first CIR is truncated, and the matrix is ​​formed. [Number 163] It further includes a second truncation unit for obtaining a processed first CIR represented by The apparatus according to claim 31, characterized in that M is a value less than N.

33. The aforementioned matrix [Number 164] The system further includes a second transformation unit for performing row and column transformations to obtain a CIR image, The aforementioned CIR image is a matrix. [Number 165] The apparatus according to claim 32, characterized in that it is represented as such, where u is the number of sampling points corresponding to the processed first CIR, and 2u × v + 1 represents the number of points in the dataset.

34. The apparatus according to any one of claims 27 to 33, characterized in that the positioning model includes a convolutional neural network (CNN) model and / or a K-nearest neighbor (KNN) model.

35. A training device for a positioning model, comprising: a memory for storing a program; and a processor for calling a program in the memory and causing the training device to execute the method according to any one of claims 1 to 9.

36. A positioning device, comprising a memory for storing a program and a processor for calling a program in the memory and causing the device to perform the method according to any one of claims 10 to 17.

37. A device comprising a processor that calls a program from memory and causes the device to execute the method described in any one of claims 1 to 17.

38. A chip characterized by including a processor that calls a program from memory and causes a device on which the chip is installed to execute the method according to any one of claims 1 to 17.

39. A computer-readable storage medium characterized in that it stores a program that causes a computer to execute the method described in any one of claims 1 to 17.

40. A computer program product characterized by including a program that causes a computer to execute the method described in any one of claims 1 to 17.

41. A computer program characterized by causing a computer to execute the method described in any one of claims 1 to 17.

Images