Indoor positioning method and device, electronic equipment, storage medium and program product
Through the spatiotemporal neural network model and position prediction model, the spatial position data and signal feature data are processed, and the problems of poor indoor positioning accuracy and high cost in the prior art are solved, thereby achieving high-precision indoor positioning.
Patent Information
- Application Number
- CN202411201072.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-29
- Publication Date
- 2024-12-17
AI Technical Summary
The existing indoor positioning technology based on mobile communication networks has poor positioning accuracy and requires additional positioning equipment to be deployed, which is costly to high manufacturing operation and maintenance.
The spatiotemporal neural network model and position prediction model are used to obtain the spatial position data of the device to be located and the signal characteristic data based on the time series to be determined to improve positioning accuracy.
It improves the accuracy of indoor positioning, reduces operation and maintenance costs, and reduces dependence on additional positioning equipment.
Smart Images
Figure CN119148050A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of wireless technology, and in particular to an indoor positioning method, device, electronic device, storage medium and program product. Background Art
[0002] Indoor precise positioning has always been a key research direction in the industry. Existing indoor positioning technologies outside of mobile communication network technologies include Bluetooth positioning, Ultra Wide Band (UWB) positioning, and WiFi fingerprint positioning. However, their shortcomings are that they require additional deployment of positioning equipment and have high manufacturing and operation costs. Therefore, in order to reduce costs, indoor positioning can be achieved based on mobile communication networks.
[0003] Existing indoor positioning technologies based on mobile communication networks mainly include wireless signal fingerprint positioning, Time Difference of Arrival (TDOA), Multi-round trip time (Multi-RTT), etc. However, their shortcomings are that the location information is obtained based on fingerprint matching or triangulation calculation, resulting in poor positioning accuracy. Summary of the invention
[0004] The present disclosure is proposed in view of the above problems. The present disclosure provides an indoor positioning method, device, electronic device, storage medium and program product.
[0005] According to one aspect of the present disclosure, there is provided an indoor positioning method, comprising:
[0006] Acquire positioning feature data corresponding to the device to be positioned, the positioning feature data including spatial position data and signal feature data based on time series, the spatial position data including the spacing data between the service TRP and each adjacent TRP, the signal feature data including the positioning signal feature data of the service TRP and the signal strength data of the adjacent TRP;
[0007] The location information corresponding to the device to be located is determined according to the positioning feature data and the positioning model corresponding to the device to be located, wherein the positioning model includes a spatiotemporal neural network model for outputting spatiotemporal feature data using the positioning feature data and a location prediction model for outputting location information using the spatiotemporal feature data, and the positioning model is trained based on a sample data set, and each sample data of the sample data set includes the positioning feature data of the sample device and a corresponding location information label.
[0008] In addition, according to an indoor positioning method according to one aspect of the present disclosure, the positioning signal characteristic data of the service TRP includes: angle information data, time information data, signal strength data, signal quality data and power change data.
[0009] In addition, according to an indoor positioning method in one aspect of the present disclosure, the positioning signal characteristic data of the service TRP further includes: channel characteristic data;
[0010] The angle data includes a horizontal arrival angle, a horizontal departure angle, a vertical arrival angle, and a vertical departure angle;
[0011] The time data includes a timing advance and a difference between uplink and downlink timing advances;
[0012] The power change data includes transmission and reception power change values;
[0013] The channel characteristic data includes channel impulse response and power delay spectrum.
[0014] In addition, according to an indoor positioning method according to one aspect of the present disclosure, the spatial position data includes a static space-time graph, the static space-time graph includes an adjacency matrix of a service TRP, and the adjacency matrix of the service TRP contains information on the spacing data between the service TRP and each adjacent TRP.
[0015] In addition, according to an indoor positioning method in one aspect of the present disclosure, the spatiotemporal neural network model is a spatiotemporal graph neural network model, a spatiotemporal convolutional neural network model, or a combined neural network model based on a long short-term memory network and a convolutional neural network;
[0016] The location prediction model is a gradient boosting decision tree model, a random forest model or a support vector machine model.
[0017] In addition, according to an indoor positioning method according to one aspect of the present disclosure, the positioning model corresponding to the device to be positioned is obtained by training based on a sample data set of each target sample device, and the target sample device is a sample device that uses the TRP of the target area as the service TRP, and the target area is a pre-set area corresponding to the service TRP of the device to be positioned and its adjacent TPR.
[0018] In addition, according to an indoor positioning method in one aspect of the present disclosure, before determining the location information corresponding to the device to be positioned according to the positioning feature data and the positioning model corresponding to the device to be positioned, the method further includes:
[0019] Receive a model association TRP operation from an operation and maintenance center, where the model association TRP operation is used to indicate association between the positioning model corresponding to the device to be positioned and the TRP of the target area;
[0020] In response to the model association TRP operation, a model association TRP request is sent to the LMF;
[0021] Receive the model association TRP response returned by the LMF, and if the association is allowed, set the association relationship between the positioning model corresponding to the device to be positioned and the TRP of the target area, wherein the association relationship is used to determine the positioning model corresponding to the device to be positioned, and is used to determine the sample data set of the target sample device;
[0022] Feedback the response of the model-associated TRP operation to the operation maintenance center.
[0023] In addition, according to an indoor positioning method according to one aspect of the present disclosure, the location information is a grid location.
[0024] According to another aspect of the present disclosure, there is provided an indoor positioning device, comprising:
[0025] A data acquisition module is used to acquire positioning feature data corresponding to the device to be positioned, wherein the positioning feature data includes spatial position data and signal feature data based on a time series, wherein the spatial position data includes the spacing data between the service TRP and each adjacent TRP, and the signal feature data includes the positioning signal feature data of the service TRP and the signal strength data of the adjacent TRP;
[0026] A position determination module is used to determine the position information corresponding to the device to be located based on the positioning feature data and the positioning model corresponding to the device to be located, wherein the positioning model includes a spatiotemporal neural network model for outputting spatiotemporal feature data using the positioning feature data and a position prediction model for outputting position information using the spatiotemporal feature data, and the positioning model is trained based on a sample data set, and each sample data of the sample data set includes the positioning feature data of the sample device and the corresponding position information label.
[0027] According to another aspect of the present disclosure, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of any of the above methods.
[0028] According to another aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program / instruction is stored. When the computer program / instruction is executed by a processor, the steps of any of the above-mentioned methods are implemented.
[0029] According to another aspect of the present disclosure, a computer program product is provided, including a computer program / instruction, which implements the steps of any of the above methods when executed by a processor.
[0030] As will be described in detail below, according to the indoor positioning method, device, electronic device, storage medium and program product of the embodiment of the present disclosure, the spatiotemporal feature data is determined based on the positioning feature data using a spatiotemporal neural network model, and the location information is determined based on the spatiotemporal feature data using a location prediction model. Among them, the positioning feature data includes spatial position data (including the spacing data between the service TRP and each adjacent TRP) and signal feature data based on a time series (the positioning signal feature data of the service TRP and the signal strength data of the adjacent TRP), which contains spatial information and time information, so that the spatiotemporal neural network model can obtain spatiotemporal feature data that can more accurately reflect the location of the device based on the positioning feature data, and then based on the spatiotemporal feature data, the location prediction model can be used to obtain accurate positioning information, thereby improving the positioning accuracy.
[0031] It is to be understood that both the foregoing general description and the following detailed description are exemplary, and are intended to provide further explanation of the technology as claimed. BRIEF DESCRIPTION OF THE DRAWINGS
[0032] The above and other purposes, features and advantages of the present disclosure will become more apparent by describing the embodiments of the present disclosure in more detail in conjunction with the accompanying drawings. The accompanying drawings are used to provide a further understanding of the embodiments of the present disclosure and constitute a part of the specification. Together with the embodiments of the present disclosure, they are used to explain the present disclosure and do not constitute a limitation of the present disclosure. In the accompanying drawings, the same reference numerals generally represent the same components or steps.
[0033] Figure 1 is a flowchart illustrating an indoor positioning method according to an embodiment of the present disclosure.
[0034] Figure 2 2 is a diagram illustrating a positioning architecture for applying an indoor positioning method according to an embodiment of the present disclosure.
[0035] Figure 3 Schematic diagram of TPR arrangement in the indoor positioning method according to an embodiment of the present disclosure.
[0036] Figure 4 It is a flowchart further illustrating the association between the model and TPR in the indoor positioning method of the embodiment of the present disclosure.
[0037] Figure 5 It is a schematic diagram of the signaling process for further illustrating the model selection and the association between the model and TRP in the indoor positioning method of an embodiment of the present disclosure.
[0038] Figure 6 It is a schematic diagram of the signaling process of training, deploying, and updating the model in the indoor positioning method of an embodiment of the present disclosure.
[0039] Figure 7 It is a signaling flow diagram further illustrating the indoor positioning method initiated by the LMF according to an embodiment of the present disclosure.
[0040] Figure 8 It is a signaling flow diagram further illustrating the indoor positioning method initiated by the UE in an embodiment of the present disclosure.
[0041] Fig. 9 is a block diagram illustrating an indoor positioning device according to an embodiment of the present disclosure.
[0042] Fig.10 is a schematic diagram illustrating a computer program product according to an embodiment of the present disclosure.
[0043] Fig.11 is a hardware block diagram illustrating an electronic device according to an embodiment of the present disclosure. DETAILED DESCRIPTION
[0044] In order to make the purpose, technical solution and advantages of the present disclosure more obvious, the exemplary embodiments according to the present disclosure will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all the embodiments of the present disclosure, and it should be understood that the present disclosure is not limited to the exemplary embodiments described here.
[0045] See also Figure 1 , an indoor positioning method, comprising:
[0046] S101, obtain the positioning feature data corresponding to the device to be positioned, the positioning feature data includes spatial position data and signal feature data based on time series, the spatial position data includes the distance data between the service TRP and each adjacent TRP, the signal feature data includes the positioning signal feature data of the service TRP and the signal strength data of the adjacent TRP.
[0047] See also Figure 2The indoor positioning method implemented in the present disclosure can be applied to the positioning architecture of the mobile communication network. The main network functions of the positioning-related core network are the access and mobility management function (AMF) and the location management function (LMF). Among them, LMF manages all processes related to the positioning of the user equipment (UE), including the selection of positioning methods, resource scheduling, global collaboration, and is also responsible for broadcasting positioning assistance information to the UE. In the access network, an architecture in which the central unit (CU) and the distributed unit (DU) are separated is adopted, and indoor signals are sent and received through the transmit-receive point (TRP). Steps S101 to S102 of the indoor positioning method can be executed in the central unit CU to realize the AI positioning function.
[0048] In step S101 of the embodiment of the present disclosure, the positioning feature data corresponding to the device to be positioned includes the spatial position data corresponding to the device to be positioned and the signal feature data based on the time series corresponding to the device to be positioned. Specifically, the spatial position data corresponding to the device to be positioned includes the spacing data between the service TRP corresponding to the device to be positioned and each adjacent TRP of the service TRP, which characterizes the spatial association relationship between the feature data of the service TRP and the adjacent TRP. In one embodiment, the area can be pre-set, and each set area is used as the adjacent TRP in each set area. That is to say, after determining the service TRP, according to the set area corresponding to the service TRP, the other TRPs in the set area except the server TRP are the adjacent TRPs of the service TRP. It should be understood that all TRPs in the set area can be formed into a binding relationship, and the adjacent TRP is directly determined by the binding relationship. Of course, the embodiment of the present disclosure can also use other methods to achieve the determination of adjacent TRPs. Specifically, the signal characteristic data based on the time series corresponding to the device to be located includes the positioning signal characteristic data of the service TRP based on the time series and the signal strength data of the adjacent TRP. The positioning signal characteristic data of the service TRP and the signal strength data of the adjacent TRP can reflect the time relationship and the signal strength data can further reflect the spatial relationship. In conjunction with the distance data, the spatiotemporal characteristic data obtained based on the positioning characteristic data can more accurately reflect the device position, thereby improving the positioning accuracy. The positioning signal characteristic data of the service TRP may include angle information data, time information data, signal strength data, signal quality data and power change data. In order to improve the positioning accuracy, the positioning signal characteristic data of the service TRP may also include channel characteristic data.
[0049] The angle data may include the horizontal angle of arrival AoA p (Horizontal angle of arrival between service TRP and equipment antenna), horizontal angle of departure AoD p (Horizontal departure angle between service TRP and equipment antenna), vertical angle of arrival ZoA p (Service TRP and the vertical angle of arrival of the mobile user's antenna) and vertical departure angle ZoD p (Service TRP and vertical departure angle of the mobile user's antenna); time data may include time advance (timing advance between signal source and mobile user) and uplink and downlink timing advance difference Power change data can include transmit and receive power change values The channel characteristic data may include the channel impulse response CIR (a vector used to characterize the strength and phase of multipath signals under different delays) and the power delay profile PDP (a quantity used to characterize the relationship between the power and arrival time delay of the signal received by the receiving end in the wireless channel). The signal strength data of the service TRP may include the service TRP signal strength value Rsrp p , the signal quality data of the service TRP may include the service TRP signal quality value Sinr p , the signal strength data of the adjacent TRP may include the signal strength value Rsrp of the adjacent TRP q .
[0050] In one embodiment, the spatial location data includes a static spatiotemporal graph, which includes an adjacency matrix of a service TRP, and the adjacency matrix of a service TRP includes information about the spacing data between the service TRP and each adjacent TRP. The static spatiotemporal graph represents the spatial association relationship of the positioning feature data. The static spatiotemporal graph can be represented as G T =(V,E T ,A T ), where V is the TRP (service TRP and adjacent TRP) distributed in the room, E represents all edge sets, A represents the adjacency matrix, T represents the service TRP, and the distance between TRPs in the adjacency matrix is the physical distance between two TRPs.
[0051] S102, determining the location information corresponding to the device to be located according to the positioning feature data and the positioning model corresponding to the device to be located, wherein the positioning model includes a spatiotemporal neural network model for outputting spatiotemporal feature data using the positioning feature data and a location prediction model for outputting location information using the spatiotemporal feature data, and the positioning model is trained based on a sample data set, and each sample data in the sample data set includes the positioning feature data of the sample device and a corresponding location information label.
[0052] In step S102 of the embodiment of the present disclosure, a spatiotemporal neural network model can be selected according to actual needs, for example, a spatiotemporal graph neural network (STGNN) model, a spatiotemporal convolutional neural network model, or a combined neural network model based on a long short-term memory network and a convolutional neural network.
[0053] In one embodiment, a spatiotemporal graph neural network model STGNN is used as the spatiotemporal neural network model. The advantage of using STGNN as a spatiotemporal feature extraction model is that it can simultaneously model data in both time and space dimensions, process graph structures through graph neural networks, and automatically extract complex spatiotemporal features, which is particularly suitable for capturing dynamic environments and long-term dependencies, and is suitable for complex indoor positioning scenarios.
[0054] Exemplarily, the positioning feature data corresponding to the device to be positioned includes the spatial position data G corresponding to the device to be positioned and the signal feature data X based on the time series corresponding to the device to be positioned. 1 ,...,X T , the input X of the spatiotemporal graph neural network model 1 ,...,X T is the characteristic data of the device to be located (such as the device of a mobile user) in the last T cycles, G is the static spatiotemporal graph obtained above, and output X 1+T It is the spatiotemporal feature data of the device to be located within the range of its spatiotemporal map.
[0055]
[0056] The spatiotemporal feature data output by the model can be used to input into a location prediction model to obtain predicted location information. Similarly, based on the model, spatiotemporal feature data for training can be output, and the spatiotemporal feature data and corresponding location tags can be used to train a location prediction model.
[0057] It should be understood that the spatiotemporal neural network model is a pre-trained model. The model can be trained based on the associated indoor distributed TRP. Accordingly, when the device using the associated indoor distributed TRP as the service TRP needs to be located, the spatiotemporal neural network model can be used to obtain spatiotemporal feature data.
[0058] In step S102 of the embodiment of the present disclosure, a location prediction model may be selected according to actual needs, for example, a gradient boosting decision tree model, a random forest model, or a support vector machine model.
[0059] In one embodiment, the gradient boosting decision tree model LightGBM is used as the location prediction model. The advantages of using LightGBM as a location prediction model are its efficient training speed, low memory usage, and ability to process high-dimensional sparse data. At the same time, it can provide high-precision prediction results through parallel learning and integration of multiple decision trees, which is very suitable for real-time location prediction tasks in large-scale and complex scenarios.
[0060] In one embodiment, the positioning model corresponding to the device to be positioned is trained based on the sample data set of each target sample device. The target sample device is a sample device that uses the TRP of the target area as the service TRP. The target area is the area corresponding to the service TRP and its adjacent TRP of the device to be positioned that is preset. In other words, the sample data trained by the positioning model corresponding to the device to be positioned corresponds to the positioning feature data corresponding to the device to be positioned. The positioning model trained in this way has a higher positioning effect for the positioning scene where the device to be positioned is located.
[0061] In an actual application scenario, you can set the associated TRP of each positioning model. For example, you can use each indoor TRP as the associated TRP of a positioning model, or you can split an indoor TRP into several regional TRPs, and use each regional TRP as the associated TRP of a positioning model. For example, TRP1, TRP2, and TRP3 are associated with model l1, and TRP4, TRP5, and TRP6 are associated with model2. Figure 3 For example, a factory building has a length of L and a width of W. 18 TRPs are arranged, the interval between two TRPs is D, and the boundary distance is D / 2. All TRPs can be bound to a positioning model, and specific training can be performed on the model to achieve the function of indoor positioning. It can also be divided into binding TRP0-8 and TRP9-17 to two positioning models respectively.
[0062] In one embodiment, the location information is a grid location. The grid location is represented by a grid code (e.g., a number). When real-time positioning of the UE is performed, the model needs to be able to obtain the input data from the air interface and perform reasoning by inputting the corresponding model. During the training data collection phase, the operation and maintenance personnel can also use the terminal to test the positioning feature data and grid code corresponding to the mobile phone to form training samples. What can be obtained during reasoning and what is obtained during data collection need to be unified.
[0063] The disclosed embodiment proposes a method for indoor distributed positioning using a double-layer stacking model, that is, a double-layer stacking model formed by a spatiotemporal neural network model (e.g., STGNN) and a position prediction model (e.g., LightGBM), wherein the spatiotemporal neural network model is used to extract features in the time domain and the spatial domain, and the position prediction model is used to determine the position. The reason for using the above two models in combination is that the position of the device to be positioned (the device of the mobile user) in the room is time-dependent, and the position features at the previous moment are strongly correlated with the position features at the next moment. On the other hand, since more TRPs are deployed in the indoor distribution, the signal features received by the user show obvious spatial characteristics (service TRP signal + adjacent TRP signal), which can well grasp the features of the time domain and the spatial domain at the same time, so that the determined spatiotemporal feature data can better reflect the device location, and the location information obtained based on the spatiotemporal feature data is more accurate.
[0064] In one embodiment, LMF manages all positioning processes, including model determination, association of the model with indoor distributed TRPs, model training, deployment, and updating, and execution of the indoor positioning method based on the disclosed embodiment, all of which are set up with signaling interaction with the LMF network element. The execution of the indoor positioning method of the disclosed embodiment involves signaling interaction with the user.
[0065] For example, in one embodiment, see Figure 4 and Figure 5 Before determining the location information corresponding to the device to be located according to the location feature data and the location model corresponding to the device to be located, the method further includes:
[0066] S401, receiving a model association TRP operation from an operation maintenance center (OMC), where the model association TRP operation is used to indicate associating a positioning model corresponding to the device to be positioned with a TRP of a target area;
[0067] See also Figure 5 Before this step, the model selection operation of the operation and maintenance center can also be received, the model selection operation can be responded to, and a model selection request can be initiated to the LMF. The model selection request includes the selected model type, current base station information and other information. The request is sent by the CU to the AMF, and further to the LMF. The LMF replies to the CU with a model selection response. The model selection response includes a unique identification of the positioning model, which is first sent to the AMF, and further sent from the AMF to the CU. If the LMF does not allow model selection, a model selection rejection message is sent to the CU, and the CU feeds back the OMC model selection response.
[0068] S402, responding to the model association TRP operation, sending a model association TRP request to the LMF.
[0069] The model association TRP operation triggers the central unit CU to initiate a model association TRP request to the LMF. The request message includes the unique identifier of the current positioning model, the bound TRP, etc. The request message is sent by the CU to the AMF and further to the LMF. The LMF replies to the CU model association TRP response. If the LMF does not allow model selection, it sends a model selection rejection message to the CU.
[0070] S403, receiving the model association TRP response returned by LMF, and if association is allowed, setting the association relationship between the positioning model corresponding to the device to be positioned and the TRP of the target area, wherein the association relationship is used to determine the positioning model corresponding to the device to be positioned, and to determine the sample data set of the target sample device.
[0071] S404: Feedback the response status of the model-associated TRP operation to the operation and maintenance center.
[0072] See also Figure 6 ,In the model training, deployment, and update signaling processes, similar processes can also be adopted, all of which are triggered by the OMC’s operations. OMC sends the model training operation to CL, CU responds to the model training operation, and initiates a model training request to LMF, which is sent by CU to AMF, and further to LMF, and LMF replies to CU whether the model training response allows the execution of the relevant operations.
[0073] The model training is implemented on the OMC. The operator imports the collected training data samples into the CU where the corresponding positioning function model is located through the OMC interface and triggers the model training. The AI modeling is completed automatically on the CU side. When the test accuracy of the positioning model reaches the threshold (such as accuracy within ±2 grids), the AI positioning model is deployed on the CU. The TRP associated with the model can now support the AI-based positioning method. When new training data is collected, the above steps can be repeated to update the model.
[0074] In a specific application scenario, see Figure 7 , LMF initiates an AI-based positioning request, which is transparently transmitted to CU through AMF. The CU side is responsible for the corresponding positioning wireless resource allocation. The UE sends a positioning response to LMF, which is sent to CU and further transmitted to AMF and LMF. LMF sends a measurement request for the positioning feature data of the positioning model to CU; CU and UE perform the measurement process of the positioning feature data of the positioning model. After completing the measurement of the positioning feature data, CU inputs the positioning feature data into the model for inference calculation to obtain the UE's location information. CU replies the location information response to LMF.
[0075] In a specific application scenario, see Figure 8, UE initiates a positioning service request, which is sent to LMF through AMF. LMF responds to the AI-based positioning service request and initiates an AI-based positioning response, which is sent to CU through AMF. The CU side is responsible for the corresponding positioning wireless resource allocation. The UE sends a positioning response to LMF, which is sent to CU and further transmitted to AMF and LMF. LMF sends a measurement request for the positioning feature data of the positioning model to CU; CU and UE perform the measurement process of the positioning feature data of the positioning model. After completing the measurement of the positioning feature data, CU inputs the data into the model for inference calculation to obtain the UE location information. CU replies the location information response to the UE.
[0076] The indoor positioning method of the embodiment of the present disclosure is based on AI positioning. When the user needs to use the indoor positioning method of the embodiment of the present disclosure, an AI-based positioning request can be initiated to indicate that the indoor positioning method of the embodiment of the present disclosure is used to implement positioning. The indoor positioning method is placed under the current mobile communication network positioning framework and uniformly managed through LMF; the configuration, training, deployment, reasoning methods and processes of the positioning model are used to realize the indoor positioning function based on AI; the signaling process of each operation of the positioning function based on AI integrates the positioning of AI into the overall signaling process of the positioning function of the mobile communication network.
[0077] See also Fig. 9 The exemplary embodiment of the present disclosure further provides an indoor positioning device, which includes:
[0078] The data acquisition module 901 is used to acquire the positioning feature data corresponding to the device to be positioned, the positioning feature data includes spatial position data and signal feature data based on time series, the spatial position data includes the spacing data between the service TRP and each adjacent TRP, and the signal feature data includes the positioning signal feature data of the service TRP and the signal strength data of the adjacent TRP;
[0079] The location determination module 902 is used to determine the location information corresponding to the device to be located based on the location feature data and the location model corresponding to the device to be located, wherein the location model includes a spatiotemporal neural network model for outputting spatiotemporal feature data using the location feature data and a location prediction model for outputting location information using the spatiotemporal feature data. The location model is trained based on a sample data set, and each sample data in the sample data set includes the location feature data of the sample device and a corresponding location information label.
[0080] In one embodiment, the positioning signal characteristic data of the service TRP includes: angle information data, time information data, signal strength data, signal quality data and power change data.
[0081] In one embodiment, the positioning signal characteristic data of the service TRP also includes: channel characteristic data.
[0082] In one embodiment, the angle data includes a horizontal angle of arrival, a horizontal angle of departure, a vertical angle of arrival, and a vertical angle of departure.
[0083] In one embodiment, the time data includes a timing advance and a difference between uplink and downlink timing advances.
[0084] In one embodiment, the power variation data includes transmit and receive power variation values.
[0085] In one embodiment, the channel characteristic data includes a channel impulse response and a power delay spectrum.
[0086] In one embodiment, the spatial location data includes a static space-time graph, which includes an adjacency matrix of the service TRP, and the adjacency matrix of the service TRP contains information on the spacing data between the service TRP and each adjacent TRP.
[0087] In one embodiment, the spatiotemporal neural network model is a spatiotemporal graph neural network model, a spatiotemporal convolutional neural network model, or a combined neural network model based on a long short-term memory network and a convolutional neural network.
[0088] In one embodiment, the location prediction model is a gradient boosting decision tree model, a random forest model, or a support vector machine model.
[0089] In one embodiment, the positioning model corresponding to the device to be positioned is trained based on the sample data sets of each target sample device. The target sample device is a sample device that uses the TRP of the target area as the service TRP. The target area is the area corresponding to the pre-set service TRP of the device to be positioned and its adjacent TRP.
[0090] In one embodiment, a correlation model is further included, for:
[0091] Receive a model association TRP operation from the operation and maintenance center, where the model association TRP operation is used to indicate the association between the positioning model corresponding to the device to be positioned and the TRP of the target area;
[0092] In response to the model association TRP operation, a model association TRP request is sent to LMF;
[0093] Receive the model association TRP response returned by LMF, and if the association is allowed, set the association relationship between the positioning model corresponding to the device to be positioned and the TRP of the target area, wherein the association relationship is used to determine the positioning model corresponding to the device to be positioned, and to determine the sample data set of the target sample device;
[0094] Feedback to the operation and maintenance center on the response of the model-associated TRP operation.
[0095] In one embodiment, the position information is a grid position.
[0096] The exemplary embodiment of the present disclosure also provides an electronic device, comprising: at least one processor; and a memory connected to the at least one processor in communication. The memory stores a computer program that can be executed by the at least one processor, and the computer program is used to cause the electronic device to perform the method according to the embodiment of the present disclosure when executed by the at least one processor.
[0097] Exemplary embodiments of the present disclosure also provide a non-transitory computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor of a computer, is used to cause the computer to execute the method according to an embodiment of the present disclosure.
[0098] refer to Fig.10 The exemplary embodiment of the present disclosure further provides a computer program product 1000, including a computer program 1001, wherein when the computer program is executed by a processor of a computer, it is used to enable the computer to perform the method according to the embodiment of the present disclosure.
[0099] refer to Fig.11 , a block diagram of an electronic device 1100 that can be used as a server or client of the present disclosure will now be described, which is an example of a hardware device that can be applied to various aspects of the present disclosure. The electronic device is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are only examples and are not intended to limit the implementation of the present disclosure described and / or required herein.
[0100] The electronic device 1100 includes a computing unit 1101, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 1102 or a computer program loaded from a storage unit 1108 into a random access memory (RAM) 1103. In the RAM 1103, various programs and data required for the operation of the device can also be stored. The computing unit 1101, the ROM 1102, and the RAM 1103 are connected to each other via a bus 1104. An input / output (I / O) interface 1105 is also connected to the bus 1104.
[0101] Multiple components in the electronic device 1100 are connected to the I / O interface 1105, including: an input unit 1106, an output unit 1107, a storage unit 1108, and a communication unit 1109. The input unit 1106 can be any type of device that can input information to the electronic device 1100, and the input unit 1106 can receive input digital or character information, and generate key signal input related to user settings and / or function control of the electronic device. The output unit 1107 can be any type of device that can present information, and can include but is not limited to a display, a speaker, a video / audio output terminal, a vibrator, and / or a printer. The storage unit 1108 can include but is not limited to a disk, an optical disk. The communication unit 1109 allows the electronic device 1100 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks, and can include but is not limited to a modem, a network card, an infrared communication device, a wireless communication transceiver, and / or a chipset, such as a Bluetooth™ device, a WiFi device, a WiMax device, a cellular communication device, and / or the like.
[0102] The computing unit 1101 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 1101 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 1101 performs the various methods and processes described above. For example, in some embodiments, the method of the embodiment of the present disclosure may be implemented as a computer software program, which is tangibly included in a machine-readable medium, such as a storage unit 1108. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 1100 via the ROM 1102 and / or the communication unit 1109. In some embodiments, the computing unit 1101 may be configured to perform the method of the embodiment of the present disclosure in any other appropriate manner (e.g., by means of firmware).
[0103] In the above, the indoor positioning method, device, electronic device, storage medium and program product according to the embodiments of the present disclosure are described with reference to the accompanying drawings. According to the indoor positioning method, device, electronic device, storage medium and program product according to the embodiments of the present disclosure, the spatiotemporal feature data is determined based on the positioning feature data using a spatiotemporal neural network model, and the location information is determined based on the spatiotemporal feature data using a location prediction model. Among them, the positioning feature data includes spatial location data (including the spacing data between the service TRP and each adjacent TRP) and signal feature data based on a time series (positioning signal feature data of the service TRP and signal strength data of the adjacent TRP), which contains spatial information and time information, so that the spatiotemporal neural network model can obtain spatiotemporal feature data that can more accurately reflect the location of the device based on the positioning feature data, and then based on the spatiotemporal feature data, the location prediction model can be used to obtain accurate positioning information, thereby improving the positioning accuracy.
[0104] The basic principles of the present disclosure are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present disclosure are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present disclosure. In addition, the specific details disclosed above are only for the purpose of illustration and ease of understanding, and are not limitations. The above details do not limit the present disclosure to the necessity of adopting the above specific details to be implemented.
[0105] The block diagrams of the devices, apparatuses, equipment, and systems involved in the present disclosure are only illustrative examples and are not intended to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagrams. As will be appreciated by those skilled in the art, these devices, apparatuses, equipment, and systems can be connected, arranged, and configured in any manner. Words such as "including", "comprising", "having", etc. are open words, referring to "including but not limited to", and can be used interchangeably with them. The words "or" and "and" used here refer to the words "and / or" and can be used interchangeably with them, unless the context clearly indicates otherwise. The word "such as" used here refers to the phrase "such as but not limited to", and can be used interchangeably with it.
[0106] Additionally, as used herein, "or" used in a list of items beginning with "at least one" indicates a separate list, so that, for example, a list of "at least one of A, B, or C" means A or B or C, or AB or AC or BC, or ABC (i.e., A and B and C). Furthermore, the word "exemplary" does not mean that the example described is preferred or better than other examples.
[0107] It should also be noted that in the system and method of the present disclosure, each component or each step can be decomposed and / or recombined. Such decomposition and / or recombination should be regarded as equivalent solutions of the present disclosure.
[0108] Various changes, substitutions, and modifications of the techniques described herein may be made without departing from the teachings defined by the appended claims. Furthermore, the scope of the claims of the present disclosure is not limited to the specific aspects of the processes, machines, manufactures, compositions of events, means, methods, and actions described above. Currently existing or later to be developed processes, machines, manufactures, compositions of events, means, methods, or actions that perform substantially the same functions or achieve substantially the same results as the corresponding aspects described herein may be utilized. Thus, the appended claims include such processes, machines, manufactures, compositions of events, means, methods, or actions within their scope.
[0109] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present disclosure. Therefore, the present disclosure is not intended to be limited to the aspects shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.
[0110] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present disclosure to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.
Claims
1. An indoor positioning method, It is characterized in that include: Acquire positioning feature data corresponding to the device to be positioned, the positioning feature data including spatial position data and signal feature data based on time series, the spatial position data including the spacing data between the service TRP and each adjacent TRP, the signal feature data including the positioning signal feature data of the service TRP and the signal strength data of the adjacent TRP; The location information corresponding to the device to be located is determined according to the positioning feature data and the positioning model corresponding to the device to be located, wherein the positioning model includes a spatiotemporal neural network model for outputting spatiotemporal feature data using the positioning feature data and a location prediction model for outputting location information using the spatiotemporal feature data, and the positioning model is trained based on a sample data set, and each sample data of the sample data set includes the positioning feature data of the sample device and a corresponding location information label.
2. The method according to claim 1, It is characterized in that The positioning signal characteristic data of the service TRP includes: angle information data, time information data, signal strength data, signal quality data and power change data.
3. The method according to claim 2, It is characterized in that The positioning signal characteristic data of the service TRP also includes: channel characteristic data; The angle data includes a horizontal arrival angle, a horizontal departure angle, a vertical arrival angle, and a vertical departure angle; The time data includes a timing advance and a difference between uplink and downlink timing advances; The power change data includes transmission and reception power change values; The channel characteristic data includes channel impulse response and power delay spectrum.
4. The method according to claim 1, It is characterized in that The spatial location data includes a static space-time graph, which includes an adjacency matrix of the service TRP, and the adjacency matrix of the service TRP contains information on the spacing data between the service TRP and each adjacent TRP.
5. The method according to claim 1, It is characterized in that The spatiotemporal neural network model is a spatiotemporal graph neural network model, a spatiotemporal convolutional neural network model, or a combined neural network model based on a long short-term memory network and a convolutional neural network; The location prediction model is a gradient boosting decision tree model, a random forest model or a support vector machine model.
6. The method according to any one of claims 1 to 5, It is characterized in that The positioning model corresponding to the device to be positioned is trained based on the sample data set of each target sample device. The target sample device is a sample device that uses the TRP of the target area as the service TRP. The target area is the area corresponding to the pre-set service TRP of the device to be positioned and its adjacent TRP.
7. The method according to claim 6, It is characterized in that Before determining the location information corresponding to the device to be located according to the positioning feature data and the positioning model corresponding to the device to be located, the method further includes: Receive a model association TRP operation from an operation and maintenance center, where the model association TRP operation is used to indicate association between the positioning model corresponding to the device to be positioned and the TRP of the target area; In response to the model association TRP operation, a model association TRP request is sent to the LMF; Receive the model association TRP response returned by the LMF, and if the association is allowed, set the association relationship between the positioning model corresponding to the device to be positioned and the TRP of the target area, wherein the association relationship is used to determine the positioning model corresponding to the device to be positioned, and is used to determine the sample data set of the target sample device; Feedback the response of the model-associated TRP operation to the operation maintenance center.
8. The method according to claim 1, It is characterized in that The position information is a grid position.
9. An indoor positioning device, It is characterized in that include: A data acquisition module is used to acquire positioning feature data corresponding to the device to be positioned, wherein the positioning feature data includes spatial position data and signal feature data based on a time series, wherein the spatial position data includes the spacing data between the service TRP and each adjacent TRP, and the signal feature data includes the positioning signal feature data of the service TRP and the signal strength data of the adjacent TRP; A position determination module is used to determine the position information corresponding to the device to be located based on the positioning feature data and the positioning model corresponding to the device to be located, wherein the positioning model includes a spatiotemporal neural network model for outputting spatiotemporal feature data using the positioning feature data and a position prediction model for outputting position information using the spatiotemporal feature data, and the positioning model is trained based on a sample data set, and each sample data of the sample data set includes the positioning feature data of the sample device and the corresponding position information label.
10. An electronic device comprising a memory, a processor and a computer program stored in the memory, It is characterized in that The processor executes the computer program to implement the steps of any one of claims 1 to 8.
11. A computer-readable storage medium having a computer program / instructions stored thereon, It is characterized in that When the computer program / instruction is executed by a processor, the steps of any method described in claims 1 to 8 are implemented.
12. A computer program product comprising a computer program / instructions, It is characterized in that When the computer program / instruction is executed by a processor, the steps of any method described in claims 1 to 8 are implemented.