Three-dimensional space wireless signal intensity determination method and device, medium and product

Abstract

The embodiment of the invention provides a three-dimensional space wireless signal intensity determination method and device, a medium and a product. The method comprises the steps that target position information of a mobile terminal in the moving process is determined; inputting the target position information into a pre-trained residual convolutional neural network, and outputting predicted wireless signal strength corresponding to the target position information; wherein the residual convolutional neural network at least comprises a residual processing sub-network, the residual processing sub-network at least comprises a sub-convolutional layer, and the sub-convolutional layer comprises two sub-full-connection layers which are connected in sequence; and determining signal intensity distribution information of the target area based on the predicted wireless signal intensity corresponding to all the target position information. According to the technical scheme provided by the embodiment of the invention, the wireless signal intensity distribution information of the target area space is determined according to the residual convolutional neural network obtained through pre-training, the problem of network gradient disappearance is relieved, the network generalization ability is enhanced, and the prediction precision of the network on the wireless signal intensity distribution is improved.

Description

Technical Field

[0001] The embodiments of the present disclosure relate to the field of communications, and in particular to a method, device, medium, and product for determining wireless signal strength in three-dimensional space. Background Art

[0002] With the development of communication technology, the signal coverage and intensity distribution information in the environment have gradually become a research hotspot in the field of wireless communications.

[0003] Current methods for predicting wireless signal strength primarily rely on convolutional neural network models. However, when training convolutional neural network models to predict wireless signal strength in an environment, the network suffers from gradient vanishing and overfitting issues, as well as low prediction accuracy and weak generalization capabilities. Summary of the Invention

[0004] The disclosed embodiments provide methods, devices, media, and products for determining wireless signal strength in three-dimensional space, thereby improving network generalization capabilities, reducing gradient vanishing and overfitting issues in the network, and improving the accuracy of wireless signal strength prediction and wireless signal strength distribution.

[0005] In a first aspect, an embodiment of the present disclosure provides a method for determining wireless signal strength in a three-dimensional space, the method comprising:

[0006] Determining target location information of the mobile terminal during movement, wherein the target location information is spatial location information relative to at least one UWB base station, and the at least one UWB base station is a positioning base station deployed in a target area;

[0007] Inputting the target location information into a pre-trained residual convolutional neural network and outputting a predicted wireless signal strength corresponding to the target location information; wherein the residual convolutional neural network includes at least a residual processing subnetwork, the residual processing subnetwork includes at least a subconvolutional layer, and the subconvolutional layer includes two sequentially connected sub-fully connected layers;

[0008] Based on the predicted wireless signal strengths corresponding to all the target location information, signal strength distribution information of the target area is determined.

[0009] In a second aspect, an embodiment of the present invention further provides a device for determining three-dimensional wireless signal strength, the device comprising:

[0010] a target location information determination module, configured to determine target location information of the mobile terminal during movement, wherein the target location information is spatial location information relative to at least one UWB base station, and the at least one UWB base station is a positioning base station deployed within a target area;

[0011] A predicted wireless signal strength output module, configured to input the target location information into a pre-trained residual convolutional neural network and output the predicted wireless signal strength corresponding to the target location information; wherein the residual convolutional neural network includes at least a residual processing subnetwork, the residual processing subnetwork includes at least a subconvolutional layer, and the subconvolutional layer includes two sequentially connected sub-fully connected layers;

[0012] The wireless signal strength distribution information determination module is configured to determine the signal strength distribution information of the target area based on the predicted wireless signal strengths corresponding to all the target location information.

[0013] In a third aspect, an embodiment of the present invention further provides an electronic device, comprising:

[0014] one or more processors;

[0015] a storage device for storing one or more programs,

[0016] When the one or more programs are executed by the one or more processors, the one or more processors implement the method for determining three-dimensional wireless signal strength as described in any one of the embodiments of the present invention.

[0017] In a fourth aspect, an embodiment of the present invention further provides a storage medium comprising computer-executable instructions, which, when executed by a computer processor, are used to execute the method for determining the three-dimensional wireless signal strength as described in any one of the embodiments of the present invention.

[0018] In a fifth aspect, an embodiment of the present invention further provides a computer program product, including a computer program, characterized in that when the computer program is executed by a processor, it implements the method for determining the three-dimensional wireless signal strength as described in any one of the embodiments of the present invention.

[0019] The technical solution of the embodiment of the present disclosure is applied to a mobile terminal to determine the target location information of the mobile terminal during movement, wherein the target location information is the spatial location information relative to at least one UWB base station, and the at least one UWB base station is a positioning base station deployed in the target area. Then, the target location information is input into a pre-trained residual convolutional neural network, and the predicted wireless signal strength corresponding to the target location information is output; wherein the residual convolutional neural network includes at least a residual processing subnetwork, and the residual processing subnetwork includes at least a subconvolutional layer, and the subconvolutional layer includes two sequentially connected sub-fully connected layers. Finally, based on the predicted wireless signal strength corresponding to all target location information, the signal strength distribution information of the target area is determined. This solves the problem in the prior art of predicting the signal strength in the environment by training a convolutional neural network, that is, the model has gradient vanishing, overfitting, weak model generalization ability, and low prediction accuracy. The embodiment of the present invention determines the wireless signal strength of the target area based on the pre-trained residual convolutional neural network. If the target area covers the entire indoor area and there may be complex environmental factors within the target area, the pre-trained residual convolutional neural network can be used to obtain the wireless signal strength at any location in the entire indoor area, thereby obtaining the wireless signal strength distribution information of the target area. This improves the network's generalization ability and the accuracy of wireless signal strength prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] To more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings introduced here only illustrate some of the embodiments to be described by the present invention, and are not exhaustive. A person skilled in the art can derive other drawings based on these drawings without inventive effort.

[0021] Figure 1 This is a flow chart of a method for determining three-dimensional wireless signal strength provided by an embodiment of the present disclosure;

[0022] Figure 2 is a schematic diagram of the layout of a UWB base station, a mobile terminal, and a tag provided in an embodiment of the present disclosure;

[0023] Figure 3 is a schematic diagram of a residual convolutional neural network provided by an embodiment of the present disclosure;

[0024] Figure 4 This is a flow chart of another method for determining three-dimensional wireless signal strength provided by an embodiment of the present disclosure;

[0025] Figure 5 is a schematic diagram of a comparison of wireless signal strength results provided by an embodiment of the present disclosure;

[0026] Figure 6 A schematic structural diagram of a device for determining three-dimensional wireless signal strength provided by an embodiment of the present invention;

[0027] Figure 7 The present invention provides a schematic structural diagram of an electronic device. DETAILED DESCRIPTION

[0028] The present invention will be further described in detail below with reference to the accompanying drawings and examples. It will be understood that the specific embodiments described herein are intended only to illustrate the present invention and are not intended to limit the present invention. It should also be noted that, for ease of description, the accompanying drawings only illustrate portions relevant to the present invention, not all structures.

[0029] Before introducing the technical solutions provided by the embodiments of the present disclosure, an example application scenario can be first described. The technical solutions provided by the embodiments of the present disclosure can be applied in scenarios where the wireless signal strength corresponding to any indoor location and the indoor wireless signal strength distribution information are determined based on a pre-trained residual convolutional neural network. For example, the 5G signal strength at each location in a certain indoor area can be determined based on the pre-trained residual convolutional neural network, thereby determining the 5G signal distribution information of the certain indoor area. It should be noted that when determining the target location information based on the technical solutions provided by the embodiments of the present disclosure, it can be performed based on the mobile terminal. The determined target location information is then input into the pre-trained residual convolutional neural network to determine the wireless signal strength and wireless signal distribution information of the target location. During the execution process of determining the target location information, at least one UMB base station and a tag device are required to cooperate. Among them, the UMB base station and the tag device are used to determine the specific location information. It should also be noted that during the execution process of determining the target location information, the mobile terminal and the tag device always maintain synchronous movement. Based on the technical solution of the embodiment of the present disclosure, combined with the residual convolutional neural network to reduce the network's gradient vanishing and overfitting defects, and the advantage of enhanced network generalization ability, the prediction accuracy of wireless signal strength distribution information is improved.

[0030] Example 1

[0031] Figure 1This is a flow chart of a method for determining the three-dimensional wireless signal strength provided by an embodiment of the present disclosure. The embodiment of the present disclosure is applicable to situations where the wireless signal strength corresponding to any indoor location and the indoor wireless signal strength distribution information are determined based on a pre-trained residual convolutional neural network. The method can be executed by a signal strength determination device, which can be implemented in the form of software and / or hardware. The hardware can be a mobile electronic device, which can execute the three-dimensional wireless signal strength determination method provided by this technical solution.

[0032] like Figure 1 As shown, the method includes:

[0033] S110: Determine target location information of the mobile terminal during movement.

[0034] Among them, the mobile terminal is any mobile device that can be carried around and can communicate with a network or external system. Common mobile terminals include smart phones, tablet computers, and wearable devices such as smart watches. In an embodiment of the present invention, the mobile terminal can be a smart phone. Optionally, the movement process of the mobile terminal is usually carried out according to test requirements or experimental scenarios. The movement of the mobile terminal can be carried out by an operator holding the mobile terminal and moving it indoors along a specified path or in any direction. The mobile terminal can also be installed on an automated platform, such as an unmanned vehicle, robot or other automatic equipment, for automated movement. It should be noted that installing the mobile terminal on an automated platform to obtain target position information during movement is usually used for large-scale signal coverage verification and other research.

[0035] The target location information is spatial location information relative to at least one UWB base station, and the at least one UWB base station is a positioning base station deployed in the target area.

[0036] It should be noted that target location information refers to the current relative position coordinates of the mobile device. Because the mobile device and the tag device are always bound and move together, the target location information also refers to the current relative position coordinates of the tag device. Furthermore, in UWB positioning systems, target location information refers to the spatial position of the mobile device or tag device relative to at least one UWB base station. Target location information can be presented using the mobile device's three-dimensional coordinates (X, Y, Z), that is, the mobile device's position in space relative to the base station. A UWB base station refers to a UWB signal transceiver device deployed in a fixed location that can detect UWB signals transmitted by tag devices. UWB base stations primarily perform positioning by transmitting broadband signals. UWB signals have high accuracy, enabling good coverage and reliable measurements in high-precision positioning scenarios, such as indoor positioning. The spatial location information of a UWB base station refers to the set of coordinates of all deployed UWB base stations within the target area and is generally used to describe the specific location of each UWB base station.

[0037] It should be noted that the target area usually refers to the specific spatial range where UWB positioning coverage is required, such as a room or several floors of a building. When deploying UWB base stations, the number and layout of UWB base stations in the target area are often determined based on the required accuracy, coverage range, and UWB base station density.

[0038] Specifically, based on the required accuracy, coverage, and density of UWB base stations in the target area, an environmental layout of at least one UWB positioning base station can be established in the target area. At least one UWB base station can be installed in different corners or at different heights in the positioning area. When installing a UWB base station, you can first confirm the main activity areas of the tag device. The UWB base station layout should cover these areas and leave appropriate margins. The deployment locations of the UWB base stations should be as dispersed as possible, and the relative positions between the UWB base stations can maintain a certain geometric distribution. When deploying UWB base stations, try to choose places with unobstructed sight or fewer obstacles. If the environment is more complex, the number of UWB base stations can be increased. When there are multiple UWB base stations, time synchronization is required between the UWB base stations. After the spatial position of the UWB positioning base station is determined, all target location information of the mobile terminal during the movement is obtained relative to the spatial location information of all installed UWB base stations.

[0039] In this embodiment, when the mobile terminal moves to the target location, it receives a positioning message sent by the tag device, and determines the target location information of the mobile terminal moving to the target location based on the sending time and receiving time in the positioning message.

[0040] It should be noted that a tag device refers to a device that can continuously send pulses or data frames to communication facilities such as UWB base stations deployed in the surrounding area. In an embodiment of the present invention, the tag device can be an independent hardware device, usually with its own processor, storage and wireless communication module. For example, the tag device can be an independent hardware device that communicates and transmits information with the UWB base station through wireless signals. The tag device can also be a part integrated into other hardware devices, for example, embedded in a product or a mobile terminal. In the embodiment of the present disclosure, the main task of the tag device is to determine the current location data and send this location information data to the mobile terminal through wireless signals.

[0041] Among them, the positioning message includes the sending time of the tag device sending the message to be verified to at least one UWB base station, and the receiving time of the tag device receiving the feedback message corresponding to the message to be verified fed back by at least one UWB base station. The tag device is bound to the mobile terminal.

[0042] It should be noted that a positioning message is a data packet containing the tag's location information and sent by the tag device, received by the mobile device. A positioning message can include the timestamp of the tag device sending and receiving the message, and the timestamp data is transmitted to the mobile device via a wireless communication network. The specific content and format of positioning messages typically vary depending on different technical standards and application scenarios.

[0043] It should be noted that to obtain the positioning message sent by the tag device, the tag device first sends a verification message to at least one UWB base station. The verification message sent by the tag device to the UWB base station is typically a request message. Typical request message contents may include the tag device ID, the timestamp of the tag device's signal transmission to the UWB base station, a high-precision clock value, or signal strength information. At the moment the tag device begins or completes message transmission, its internal timing module generates or reads a high-precision clock value. This clock value serves as the transmission time. Then, after at least one UWB base station receives the verification message sent by the tag device, the UWB base station provides feedback to the tag device based on the verification message. The feedback message from the UWB base station to the tag device is typically a response message, informing the tag device of the signal strength measured by the UWB base station or other relevant information. For example, the UWB base station may send the tag device messages regarding system status, clock synchronization, calibration, or other parameters. At the moment the tag device receives the feedback message, its internal timing module generates or reads a high-precision clock value. This clock value serves as the reception time. The purpose of the tag device sending a verification message to the UWB base station and the tag device receiving the feedback message corresponding to the verification message is to obtain the target location information of the mobile terminal moving to the target location. It should also be noted that when the UWB base station receives the verification message, it needs to perform a series of verification or decoding according to the system's business logic or protocol process to confirm the validity of the verification message. After being verified by the UWB base station, the message will be officially adopted or used.

[0044] Specifically, such as Figure 2 As shown, during the movement of the mobile terminal, the tag device and the mobile terminal are always in a bound state. When the mobile terminal moves to the tag device, the tag device is also at the target location. After the tag device sends the message to be verified to all UWB base stations, the UWB base station feeds back a feedback message corresponding to the message to be verified. Then, the tag device receives the feedback messages from all UWB base stations. The tag device stores the sending time of the message to be verified sent by the tag device to at least one UWB base station, and the receiving time of the feedback message corresponding to the message to be verified fed back by the tag device from at least one UWB base station. The set of sending time and receiving time is called the positioning message of the tag device. Further, the tag device sends the positioning message to the mobile terminal. The mobile terminal determines the target location information based on the sending time and receiving time in the positioning message.

[0045] In this embodiment, the distance information between the UWB base station and the tag device is determined based on the sending time, receiving time and preset medium propagation speed in the positioning message; the target position information of the mobile terminal moving to the target position is determined based on the distance information between the UWB base station and the tag device and the spatial position information of the UWB base station.

[0046] Distance information refers to the linear distance between each UWB base station and the tag device. During the interaction between the UWB base station and the tag device, the message propagation speed, or the preset medium propagation speed, is primarily determined by the propagation speed of electromagnetic waves in the medium. For radio signals propagating in air, the speed is essentially equal to the speed of light. In air, the message propagation speed is approximately the speed of light, or approximately 299,792,458 meters per second.

[0047] It should be noted that the distance information between the UWB base station and the tag device can be determined based on the sending time, receiving time, preset medium propagation speed and distance information calculation formula in the positioning message.

[0048] The distance information calculation formula is:

[0049]

[0050] Where i represents a unique identifier assigned to each UWB base station, that is, it represents a different UWB base station. d(i) represents the distance information between each UWB base station and the sending time, receiving time, and preset medium propagation speed in the positioning message. c represents the preset medium propagation speed. i It represents the receiving time, i.e. the time when the tag device receives the signal from each UWB base station. t represents the sending time, i.e. the time when the tag device sends the signal.

[0051] It should also be noted that the spatial location information of UWB base stations refers to the relative coordinates of all UWB base stations. For example, when there are four UWB base stations, the coordinates of one UWB base station can be set to (0,0,0), and the relative spatial coordinates of the remaining three UWB base stations can be determined based on the measured distance.

[0052] Specifically, first, the sending time of the tag device sending a message to all UWB base stations at the same time is obtained. Then, according to the identifier order of the UWB base stations, the tag device obtains the signal reception time of each UWB base station feedback signal to the tag device in turn. After obtaining the signal sending time, receiving time and the preset medium propagation speed, the distance information between each UWB base station and the tag device is finally obtained according to the distance information calculation formula. After obtaining the distance information, the distance information is sent to the mobile terminal based on the tag device. After the mobile terminal obtains the spatial location information of at least one UWB base station and the distance information between the UWB base station and the tag device, the current position of the tag device can be calculated. The current position of the tag device is the target position information when the mobile terminal moves to the target position.

[0053] For example, when the signal transmission time of the tag device is t, the signal reception time when the tag device receives the signal from the first UWB base station is t1, the signal reception time when the tag device receives the signal from the second UWB base station is t2, the signal reception time when the tag device receives the signal from the third UWB base station is t3, and the signal reception time when the tag device receives the signal from the fourth UWB base station is t4, based on the signal propagation speed, it can be determined that the distance information between the first UWB base station and the signal transmitting device is d1, the distance information between the second UWB base station and the signal transmitting device is d2, the distance information between the third UWB base station and the signal transmitting device is d3, and the distance information between the fourth UWB base station and the signal transmitting device is d4. Furthermore, assuming that the spatial position information of the first UWB base station is (x0, y0, z0), the spatial position information of the second UWB base station is (x1, y1, z1), the spatial position information of the third UWB base station is (x2, y2, z2), and the spatial position information of the fourth UWB base station is (x3, y3, z3). All distance information is sent to the mobile terminal based on the signal sending device. The mobile terminal obtains the spatial location information of at least one UWB base station and the distance information between the UWB base station and the tag device. Assuming that the current position of the tag device is (x, y, z), according to as well as The current position (x, y, z) of the tag device can be obtained, that is, the target position information of the mobile terminal moving to the target position can be obtained.

[0054] S120: Input the target location information into a pre-trained residual convolutional neural network, and output the predicted wireless signal strength corresponding to the target location information.

[0055] Among them, such as Figure 3 As shown, the input X of the residual convolutional neural network is the target position information, and the output Y of the residual convolutional neural network is the predicted wireless signal strength corresponding to the target position information. Before inputting the target position information into the pre-trained residual convolutional neural network, the target position information needs to be organized into the input format required by the network. Optionally, the residual convolutional neural network usually requires two-dimensional or three-dimensional input data. If the position information is one-dimensional, appropriate conversion or expansion may be required. In an embodiment of the present invention, the target position information can be subjected to necessary preprocessing, such as normalization, standardization or other network-specific preprocessing steps, to ensure that the shape of the input target position information data is consistent with the expected shape of the network input layer.

[0056] The predicted wireless signal strength refers to the strength of the wireless signal power value received by the mobile terminal at the target location, as predicted by the residual convolutional neural network. For example, the power strength of the received wireless signal can be the power strength of a 5G signal received by a mobile phone in wireless communication.

[0057] Optionally, determining the target signal strength at the current location can evaluate the performance of different communication protocols and technologies by measuring and analyzing the wireless signal strength, which helps to optimize the transmission rate, reduce interference and improve communication reliability. In addition, when developing new antenna designs, modulation technologies or signal processing algorithms, determining the target signal strength at the current location can help verify the effectiveness and performance of the technology. Variables can also be controlled to study the impact of different environmental factors (such as wall materials, equipment interference) on signal strength, thereby providing data support for practical applications. In practical applications, measuring signal strength can not only help optimize the coverage of wireless networks and ensure stable connections in buildings or open areas, but also use signal strength measurements to plan base station locations and configurations to ensure good coverage and capacity, etc. Among them, the residual convolutional neural network includes at least a residual processing subnetwork, the residual processing subnetwork includes at least a subconvolutional layer, and the subconvolutional layer includes two sequentially connected sub-fully connected layers.

[0058] It should be noted that there are many network designs in the residual convolutional neural network, and the residual convolutional neural network can include more refined components. The residual processing subnetwork is one of the refined components in the residual convolutional neural network, and as a functional module, the residual processing subnetwork is responsible for extracting or processing the features of the input signal in the network and transmitting information through residual connections. The residual processing subnetwork includes at least a subconvolution layer. The subconvolution layer refers to a specific implementation in the residual processing subnetwork. The subconvolution layer is usually composed of a convolution layer (such as a 3*3 convolution) and is used to extract local features. The subconvolution layer includes two sequentially connected sub-fully connected layers, which means that after the convolution operation, the high-level representation capability of the features is enhanced through two fully connected layers.

[0059] Specifically, a residual convolutional neural network is first constructed. Then, the trained convolutional neural network is used to input the target location information into the trained convolutional neural network. The output is the predicted wireless signal strength corresponding to the target location information.

[0060] In this embodiment, based on the input-output relationship, the residual convolutional neural network includes an input layer, a residual processing subnetwork, a convolution layer, a pooling layer, at least two fully connected layers and an output layer in sequence. The residual processing subnetwork includes a sub-convolution layer, an activation function layer and a skip connection layer. The weight distribution layer in the sub-convolution layer is replaced by a sub-fully connected layer.

[0061] The input layer in a residual convolutional neural network (RCN) is the starting point of the network, receiving external input data and passing the raw data to subsequent layers for feature extraction and learning. The residual processing subnetwork in a RCN is the core module of the RCN, performing feature extraction through residual connections. Within each residual block, skip connections are used to add the input signal to the signal processed by the convolutional layer, helping the network train deeper. The convolutional layer in a RCN is used to extract local features of the input data. In each convolutional layer, sliding convolutions are performed on the input using a convolution kernel. The pooling layer in a RCN is used to reduce the spatial size of the feature map, thereby reducing computational effort and preventing overfitting. The pooling layer performs downsampling by selecting the maximum or average value in a specific region. The fully connected layer in a RCN is a high-level structure in the network, typically located after the convolutional and pooling layers, and is responsible for the final classification or regression task. The fully connected layer maps the previously extracted features to a higher-dimensional space and ultimately maps them to the output predictions. At least two fully connected layers can further combine the features extracted by the convolutional layer and the pooling layer to perform higher-level feature combination to improve the performance of the model; the output layer in the residual convolutional neural network is the final layer of the network, responsible for generating the final output of the network.

[0062] It should be noted that the sub-convolutional layer in the residual processing sub-network is used to process each local area of ​​the input data and perform feature abstraction. The weight distribution layer in the factor convolution layer is replaced by a sub-fully connected layer, which will make the network processing process more complicated because the fully connected layer is usually more computationally intensive than the weight distribution layer; the activation function layer in the residual processing sub-network is used to introduce nonlinear characteristics. Through the activation function, the network can learn more complex features and enhance its representation ability; the skip connection layer in the residual processing sub-network can help the network avoid the problems of gradient disappearance and gradient explosion. The skip connection layer directly adds the input signal to the output, and the skip connection layer allows the gradient to be directly back-propagated to the previous layer to maintain information flow; the weight distribution layer usually refers to the convolution kernel of the convolution layer, which controls the learning and distribution of features.

[0063] Specifically, the network structure of the residual convolutional neural network includes an input layer, a residual processing subnetwork, a convolution layer, a pooling layer, a fully connected layer, and an output layer. The key modification lies in the residual processing subnetwork, which uses subconvolution layers, activation function layers, and skip connection layers, while replacing the traditional subconvolution layer's weight distribution layer with a sub-fully connected layer. When sub-fully connected layers are used instead of weight distribution layers, the output of each layer is mapped to a higher-dimensional space, usually through matrix operations in the fully connected layer, which increases computational complexity and leads to stronger feature learning capabilities.

[0064] S130: Determine signal strength distribution information of the target area based on the predicted wireless signal strengths corresponding to all target location information.

[0065] The signal strength distribution information refers to the spatial distribution of wireless signal strength at various locations within the target area.

[0066] It should be noted that location information can be collected from multiple locations within the target area. These locations can be grid points, randomly sampled points, or locations sampled based on other reasonable sampling strategies. For each location, a pre-trained residual convolutional neural network is used to predict wireless signal strength. Optionally, the predicted wireless signal strength data can be mapped to the coordinate system of the target area, creating a matrix or grid to represent the signal strength distribution information within the target area. Furthermore, statistical metrics, such as average signal strength and signal strength variance, can be calculated to assess the quality of signal coverage in the target area.

[0067] Optionally, you can use visualization tools (such as Matplotlib, Seaborn, or Plotly) to create a heat map or contour map of the signal strength in the target area. This can help identify signal strength hotspots and blind spots in the target area. Furthermore, based on the signal strength distribution map, you can identify areas with insufficient signal coverage. Based on the identification results, you can perform further optimization, such as continuing to determine location information based on the mobile terminal and tag devices to obtain predicted wireless signal strength in blind spots.

[0068] Specifically, all target location information is fed into the trained residual convolutional neural network, which outputs the predicted wireless signal strength corresponding to all target location information. Once the predicted wireless signal strength corresponding to all target location information is obtained, the signal strength distribution information for the target area can be determined.

[0069] The technical solution of the embodiment of the present disclosure is applied to a mobile terminal to determine the target location information of the mobile terminal during movement, wherein the target location information is the spatial location information relative to at least one UWB base station, and the at least one UWB base station is a positioning base station deployed in the target area. Then, the target location information is input into a pre-trained residual convolutional neural network, and the predicted wireless signal strength corresponding to the target location information is output; wherein the residual convolutional neural network includes at least a residual processing subnetwork, and the residual processing subnetwork includes at least a subconvolutional layer, and the subconvolutional layer includes two sequentially connected sub-fully connected layers. Finally, based on the predicted wireless signal strength corresponding to all target location information, the signal strength distribution information of the target area is determined. This solves the problem in the prior art of predicting the signal strength in the environment by training a convolutional neural network, that is, the model has gradient vanishing, overfitting, weak model generalization ability, and low prediction accuracy. The embodiment of the present invention determines the signal strength of the target area based on the pre-trained residual convolutional neural network. If the target area covers the entire indoor area and there may be complex environmental factors within the target area, the pre-trained residual convolutional neural network can be used to obtain the signal strength at any location in the entire indoor area, thereby obtaining the signal strength distribution information of the target area. This improves the network's generalization ability and the accuracy of signal strength prediction.

[0070] Example 2

[0071] Figure 4 This is a flow chart of a method for determining three-dimensional wireless signal strength provided by an embodiment of the present invention. Based on the previous embodiment, a detailed description of the residual convolutional neural network trained is provided. For its specific implementation, please refer to the technical solution of this embodiment. Technical terms that are identical or corresponding to those in the above embodiment are not repeated here.

[0072] like Figure 4 As shown, the method specifically includes the following steps:

[0073] S210: Determine target location information of the mobile terminal during movement.

[0074] S220. Construct training samples for training the residual convolutional neural network.

[0075] The training samples include sample position information within the target area and measured signal strength corresponding to the sample position information.

[0076] It should be noted that in order to improve the accuracy of model training, different sample location information within the target area can be obtained, and the measured signal strength corresponding to the different sample location information can be obtained. The sum of all the different sample location information within the target area and the measured signal strength corresponding to the different sample location information constitutes the training sample. That is, the training sample includes multiple different sample location information. Among them, the sample location information refers to the data of certain specific locations within the target area. In addition, the sample location information can be determined based on the sending time, receiving time, preset medium propagation speed and spatial location information of the UWB base station in the positioning message. The measured signal strength refers to the signal strength value actually measured at the sample position. The measured signal strength corresponding to the sample location information can be used to make labels in the network learning process to help the network learn the mapping relationship from location information within the target area to signal strength.

[0077] Optionally, to improve the accuracy of the trained model, you can obtain as much and rich information as possible about the locations of different samples, and adjust the batch size, training rounds, and learning rate of the model to obtain the best prediction results.

[0078] Specifically, training samples for training the residual convolutional neural network are constructed based on the sample position information within the target area and the measured signal strength corresponding to the sample position information.

[0079] S230: Input the training sample into the residual convolutional neural network and output the signal strength to be calibrated.

[0080] The network parameters in the residual convolutional neural network to be trained are the initial parameters or the model with default parameters. The signal strength to be calibrated is the wireless signal strength output after the current different sample position information is input into the residual convolutional neural network to be trained.

[0081] Specifically, the target location information is converted into an acceptable input format for the network and then fed into the residual convolutional neural network. The measured signal strength is also fed into the residual convolutional neural network as label data. These measured signal strengths are used to supervise network learning. Furthermore, the input features can be normalized or standardized to ensure they are within a similar numerical range, which helps accelerate the training process and improve network performance. After the training samples are fed into the residual convolutional neural network, the output is the signal strength to be calibrated.

[0082] S240, performing loss processing on the measured signal strength of the training sample and the signal strength to be calibrated based on the loss function in the residual convolutional neural network, and correcting the network parameters in the residual convolutional neural network with the loss value based on the loss processing.

[0083] It should be noted that the network parameters in the residual convolutional neural network to be trained do not meet the expected requirements. Therefore, there is a certain difference between the signal strength to be calibrated output based on the network parameters at this time and the measured signal strength. Therefore, the corresponding error loss value can be determined based on the measured signal strength and the signal strength to be calibrated corresponding to each sample position information.

[0084] It should be noted that the training parameters can be set to default values ​​before training the residual convolutional neural network to be trained. When training the residual convolutional neural network to be trained, the training parameters in the model can be corrected based on the output results of the residual convolutional neural network to be trained. In other words, the residual convolutional neural network can be obtained by correcting the loss function in the residual convolutional neural network to be trained. Each sample position information has a corresponding loss value, which is determined based on the signal strength to be calibrated and the measured signal strength of each sample position information.

[0085] Specifically, after inputting each sample position information into the residual convolutional neural network to be trained, the residual convolutional neural network to be trained can obtain the signal strength to be calibrated corresponding to each sample position information. Based on the signal strength to be calibrated and the measured signal strength, the loss value corresponding to the sample position information can be determined, and the network parameters of the residual convolutional neural network to be trained can be corrected using a mean square error function.

[0086] S250. The model obtained when the loss function converges is used as a residual convolutional neural network for processing target position information.

[0087] Wherein, the loss function is a mean square error function.

[0088] Specifically, the training error of the loss function, that is, the loss parameter, can be used as a condition for detecting whether the loss function has reached convergence, such as whether the training error is less than the preset error or whether the error change trend tends to be stable, or whether the current number of iterations is equal to the preset number. If the detection reaches the convergence condition, such as the training error of the loss function is less than the preset error or the error change tends to be stable, it indicates that the training of the residual convolutional neural network to be trained is completed, and the iterative training can be stopped at this time. If it is detected that the convergence condition is not met at present, the sample position information can be further obtained to train the residual convolutional neural network to be trained until the training error of the loss function is within the preset range. When the training error of the loss function reaches convergence, the residual convolutional neural network to be trained can be used as the residual convolutional neural network to process the target position information.

[0089] S260. Verify the accuracy of the residual convolutional neural network based on multiple test samples.

[0090] The test samples are samples from the same distribution as the training samples, but the test samples are not used for network training. The test samples should include target location information and the corresponding measured signal strength. Optionally, the test samples are subjected to the same preprocessing steps as the training data, such as normalization or standardization. It should be noted that commonly used evaluation indicators for accuracy verification include mean square error, root mean square error, or mean absolute error. These indicators can quantify the difference between the predicted wireless signal strength and the measured signal strength. Optionally, a scatter plot of the measured signal strength and the predicted wireless signal strength can be drawn to intuitively observe the network's prediction performance.

[0091] Specifically, verifying the accuracy of the residual convolutional neural network is a key step in ensuring that the network performs well on unseen data. This is typically done by using multiple test samples to evaluate the network's predictive performance and verifying the accuracy of the residual convolutional neural network.

[0092] For example, the accuracy of the residual convolutional neural network is verified based on multiple test samples. Figure 5 As shown in the figure, the comparison results of the predicted wireless signal strength and the measured signal strength obtained based on the residual convolutional neural network can be displayed.

[0093] S270. When the accuracy verification result meets the preset conditions, a residual convolutional neural network for processing the target position information is obtained.

[0094] Before training and validating the network, clear evaluation criteria can be set. For example, an acceptable threshold can be set for mean squared error, root mean square error, or mean absolute error. The network is validated using a test dataset and the evaluation criteria are calculated. The calculated evaluation metrics are compared with pre-defined criteria. If the network's evaluation metrics meet or exceed the pre-defined criteria, the network can be considered to perform well on the target task. This indicates that the accuracy verification results meet the pre-defined criteria, meaning that the parameter errors are within acceptable thresholds. At this point, the network can be saved for later use.

[0095] Specifically, by setting clear preconditions and a rigorous evaluation process, we can ensure the effectiveness of the residual convolutional neural network in processing target location information. Once the model is verified and meets the expected standards, the residual convolutional neural network that processes target location information is obtained and can be used in real-world applications.

[0096] S280: Input the target location information into a pre-trained residual convolutional neural network, and output the predicted wireless signal strength corresponding to the target location information.

[0097] S290: Determine signal strength distribution information of the target area based on the predicted wireless signal strengths corresponding to all target location information.

[0098] The technical solution of the embodiment of the present disclosure is applied to a mobile terminal to determine the target position information of the mobile terminal during movement, and then construct a training sample for training the residual convolutional neural network, input the training sample into the residual convolutional neural network, and output the signal strength to be calibrated. The measured signal strength of the training sample and the signal strength to be calibrated are lost based on the loss function in the residual convolutional neural network, and the network parameters in the residual convolutional neural network are corrected based on the loss value of the loss processing. The model obtained when the loss function converges is used as the residual convolutional neural network for processing the target position information. Furthermore, the accuracy of the residual convolutional neural network is verified based on multiple test samples. When the accuracy verification result meets the preset conditions, the residual convolutional neural network for processing the target position information is obtained. Finally, the target location information is input into the pre-trained residual convolutional neural network, and the predicted wireless signal strength corresponding to the target location information is output. Based on the predicted wireless signal strength corresponding to all target location information, the signal strength distribution information of the target area is determined. According to the pre-trained residual convolutional neural network, the signal strength at any location in the entire indoor area can be obtained, and then the signal strength distribution information of the target area can be obtained, thereby achieving the effect of improving the network generalization ability and signal strength prediction accuracy.

[0099] Example 3

[0100] Figure 6 FIG. 1 is a structural diagram of a device for determining a three-dimensional wireless signal strength provided by an embodiment of the present disclosure. Figure 6 As shown, the device includes: a target location information determination module 310 , a predicted wireless signal strength output module 320 , and a wireless signal strength distribution information determination module 330 .

[0101] A target position information determination module is used to determine the target position information of the mobile terminal during the movement process, wherein the target position information is the spatial position information relative to at least one UWB base station, and the at least one UWB base station is a positioning base station deployed in the target area; a predicted wireless signal strength output module is used to input the target position information into a pre-trained residual convolutional neural network, and output the predicted wireless signal strength corresponding to the target position information; wherein the residual convolutional neural network includes at least a residual processing subnetwork, and the residual processing subnetwork includes at least a subconvolutional layer, and the subconvolutional layer includes two sequentially connected sub-fully connected layers; a wireless signal strength distribution information determination module is used to determine the signal strength distribution information of the target area based on the predicted wireless signal strength corresponding to all the target position information.

[0102] The technical solution of the embodiment of the present disclosure is applied to a mobile terminal to determine the target location information of the mobile terminal during movement, wherein the target location information is the spatial location information relative to at least one UWB base station, and the at least one UWB base station is a positioning base station deployed in the target area. Then, the target location information is input into a pre-trained residual convolutional neural network, and the predicted wireless signal strength corresponding to the target location information is output; wherein the residual convolutional neural network includes at least a residual processing subnetwork, and the residual processing subnetwork includes at least a subconvolutional layer, and the subconvolutional layer includes two sequentially connected sub-fully connected layers. Finally, based on the predicted wireless signal strength corresponding to all target location information, the signal strength distribution information of the target area is determined. This solves the problem in the prior art of predicting the signal strength in the environment by training a convolutional neural network, that is, the model has gradient vanishing, overfitting, weak model generalization ability, and low prediction accuracy. The embodiment of the present invention determines the signal strength of the target area based on the pre-trained residual convolutional neural network. If the target area covers the entire indoor area and there may be complex environmental factors within the target area, the pre-trained residual convolutional neural network can be used to obtain the signal strength at any location in the entire indoor area, thereby obtaining the signal strength distribution information of the target area. This improves the network's generalization ability and the accuracy of signal strength prediction.

[0103] On the basis of the above technical solutions, the target location information determination module 310 includes: a positioning message receiving submodule and a target location information determination submodule.

[0104] A positioning message receiving submodule is used to receive a positioning message sent by a tag device when the mobile terminal moves to a target location, wherein the positioning message includes the sending time of the tag device sending the message to be verified to the at least one UWB base station, and the receiving time of the tag device receiving a feedback message corresponding to the message to be verified fed back by the at least one UWB base station, and the tag device is bound to the mobile terminal; a target location information determination submodule is used to determine the target location information of the mobile terminal moving to the target location based on the sending time and the receiving time in the positioning message.

[0105] Based on the above technical solutions, the target location information determination submodule also includes: determining the distance information between the UWB base station and the tag device based on the sending time, the receiving time and the preset medium propagation speed in the positioning message; determining the target location information of the mobile terminal moving to the target location based on the distance information between the UWB base station and the tag device and the spatial location information of the UWB base station.

[0106] On the basis of the above technical solutions, the residual convolutional neural network includes an input layer, the residual processing subnetwork, a convolution layer, a pooling layer, at least two fully connected layers and an output layer in sequence according to the input-output relationship. The residual processing subnetwork includes a sub-convolution layer, an activation function layer and a skip connection layer, and the weight distribution layer in the sub-convolution layer is replaced by the sub-fully connected layer.

[0107] On the basis of the above technical solutions, the device further includes: a training sample construction module, a signal strength output module to be calibrated, a network parameter correction module and a residual convolutional neural network determination module.

[0108] A training sample construction module is used to construct training samples for training the residual convolutional neural network; wherein the training samples include sample position information within the target area and the measured signal strength corresponding to the sample position information; a signal strength output module to be calibrated is used to input the training samples into the residual convolutional neural network and output the signal strength to be calibrated; a network parameter correction module is used to perform loss processing on the measured signal strength of the training samples and the signal strength to be calibrated based on the loss function in the residual convolutional neural network, and to correct the network parameters in the residual convolutional neural network with the loss value based on the loss processing; a residual convolutional neural network determination module is used to use the model obtained when the loss function converges as the residual convolutional neural network for processing the target position information; wherein the loss function is a mean square error function.

[0109] Based on the above technical solutions, the device also includes: an accuracy verification module and a residual convolutional neural network determination module.

[0110] An accuracy verification module is used to verify the accuracy of the residual convolutional neural network based on multiple test samples; a residual convolutional neural network determination module is used to obtain a residual convolutional neural network for processing the target position information when the accuracy verification result meets the preset conditions.

[0111] The signal strength determination device provided in the embodiments of the present disclosure can execute the signal strength determination method provided in any embodiment of the present disclosure, and has the corresponding functional modules and beneficial effects of the execution method.

[0112] It is worth noting that the various units and modules included in the above-mentioned device are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the protection scope of the embodiments of the present disclosure.

[0113] Example 4

[0114] Figure 7 This is a schematic diagram of the structure of an electronic device provided by an embodiment of the present disclosure. Figure 7 , which shows an electronic device (eg Figure 7 The terminal device in the embodiments of the present disclosure may include, but is not limited to, a mobile terminal such as a mobile phone, a laptop computer, a digital broadcast receiver, a PDA (personal digital assistant), a PAD (tablet computer), a PMP (portable multimedia player), an in-vehicle terminal (such as an in-vehicle navigation terminal), and the like. Figure 7 The electronic device shown is only an example and should not limit the functions and scope of use of the embodiments of the present disclosure.

[0115] like Figure 7 As shown, the electronic device 500 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 501, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 502 or a program loaded from a storage device 508 into a random access memory (RAM) 503. Various programs and data required for the operation of the electronic device 500 are also stored in the RAM 503. The processing device 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An edit / output (I / O) interface 505 is also connected to the bus 504.

[0116] Typically, the following devices may be connected to the I / O interface 505: an input device 506 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 507 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 508 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 509. The communication device 509 may allow the electronic device 500 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 7 The electronic device 500 is shown with various devices, but it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed instead.

[0117] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 509, or installed from the storage device 508, or installed from the ROM 502. When the computer program is executed by the processing device 501, the above-mentioned functions defined in the method of the embodiment of the present disclosure are performed.

[0118] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0119] The electronic device provided in the embodiment of the present disclosure and the signal strength determination method provided in the above embodiment belong to the same inventive concept. For technical details not fully described in this embodiment, please refer to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0120] Example 5

[0121] An embodiment of the present disclosure provides a computer storage medium having a computer program stored thereon. When the program is executed by a processor, the signal strength determination method provided in the above embodiment is implemented.

[0122] It should be noted that the computer-readable medium mentioned above in the present disclosure may be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or component, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, device, or component. In the present disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to wires, optical cables, RF (radio frequency), etc., or any suitable combination thereof.

[0123] In some embodiments, the server can communicate using any currently known or later developed network protocol, such as HTTP (HyperText Transfer Protocol), and can be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or later developed network.

[0124] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0125] The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device:

[0126] Determining target location information of the mobile terminal during movement, wherein the target location information is spatial location information relative to at least one UWB base station, and the at least one UWB base station is a positioning base station deployed in a target area;

[0127] Inputting the target location information into a pre-trained residual convolutional neural network and outputting a predicted wireless signal strength corresponding to the target location information; wherein the residual convolutional neural network includes at least a residual processing subnetwork, the residual processing subnetwork includes at least a subconvolutional layer, and the subconvolutional layer includes two sequentially connected sub-fully connected layers;

[0128] Based on the predicted wireless signal strengths corresponding to all the target location information, signal strength distribution information of the target area is determined.

[0129] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages, or a combination thereof, including, but not limited to, object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on the remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0130] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the module, program segment, or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of the boxes in the block diagram and / or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0131] The units involved in the embodiments described in this disclosure may be implemented in software or hardware, wherein the name of a unit does not necessarily limit the unit itself.

[0132] The functions described above herein may be performed, at least in part, by one or more hardware logic components. For example, and without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chip (SOCs), complex programmable logic devices (CPLDs), and the like.

[0133] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in conjunction with an instruction execution system, device or equipment. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium can include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0134] The above description is merely a preferred embodiment of the present disclosure and an illustration of the technical principles employed. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by the specific combination of the above-mentioned technical features, but also includes other technical solutions formed by any combination of the above-mentioned technical features or their equivalents without departing from the above-mentioned disclosed concepts. For example, a technical solution formed by replacing the above-mentioned features with (but not limited to) technical features with similar functions disclosed in this disclosure.

[0135] In addition, although each operation is described in a specific order, this should not be understood as requiring these operations to be performed in the specific order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details have been included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0136] Although the subject matter has been described in language specific to structural features and / or methodological logical acts, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are merely example forms of implementing the claims.

Claims

1. A method for determining wireless signal strength in three-dimensional space, characterized in that: Applied to a mobile terminal, the method includes: Determining target location information of the mobile terminal during movement, wherein the target location information is spatial location information relative to at least one UWB base station, and the at least one UWB base station is a positioning base station deployed in a target area; Inputting the target location information into a pre-trained residual convolutional neural network and outputting a predicted wireless signal strength corresponding to the target location information; wherein the residual convolutional neural network includes at least a residual processing subnetwork, the residual processing subnetwork includes at least a subconvolutional layer, and the subconvolutional layer includes two sequentially connected sub-fully connected layers; Based on the predicted wireless signal strengths corresponding to all the target location information, signal strength distribution information of the target area is determined.

2. The method according to claim 1, characterized in that Determining the target location information of the mobile terminal during movement includes: when the mobile terminal moves to the target location, receiving a positioning message sent by a tag device, wherein the positioning message includes the sending time of the tag device sending the message to be verified to the at least one UWB base station, and the receiving time of the tag device receiving a feedback message corresponding to the message to be verified fed back by the at least one UWB base station, and the tag device is bound to the mobile terminal; The target location information of the mobile terminal moving to the target location is determined based on the sending time and the receiving time in the positioning message.

3. The method according to claim 2, characterized in that The determining, based on the sending time and the receiving time in the positioning message, target location information of the mobile terminal moving to the target location includes: Determine the distance between the UWB base station and the tag device based on the sending time, the receiving time, and a preset medium propagation speed in the positioning message; The target position information of the mobile terminal moving to the target position is determined according to the distance information between the UWB base station and the tag device and the spatial position information of the UWB base station.

4. The method according to claim 1, wherein According to the input-output relationship, the residual convolutional neural network includes an input layer, the residual processing subnetwork, a convolution layer, a pooling layer, at least two fully connected layers and an output layer in sequence. The residual processing subnetwork includes a subconvolution layer, an activation function layer and a skip connection layer. The weight distribution layer in the subconvolution layer is replaced by the sub-fully connected layer.

5. The method according to claim 1, wherein The method further comprises: Constructing training samples for training the residual convolutional neural network; wherein the training samples include sample position information within the target area and measured signal strength corresponding to the sample position information; Inputting the training samples into the residual convolutional neural network and outputting the signal strength to be calibrated; Performing loss processing on the measured signal strength of the training sample and the signal strength to be calibrated based on the loss function in the residual convolutional neural network, and correcting the network parameters in the residual convolutional neural network with the loss value based on the loss processing; The model obtained when the loss function converges is used as a residual convolutional neural network for processing the target position information; Wherein, the loss function is a mean square error function.

6. The method according to claim 5, characterized in that The method further comprises: Verifying the accuracy of the residual convolutional neural network based on multiple test samples; When the accuracy verification result meets the preset conditions, a residual convolutional neural network for processing the target position information is obtained.

7. A device for determining wireless signal strength in three-dimensional space, characterized in that: include: a target location information determination module, configured to determine target location information of the mobile terminal during movement, wherein the target location information is spatial location information relative to at least one UWB base station, and the at least one UWB base station is a positioning base station deployed within a target area; A predicted wireless signal strength output module, configured to input the target location information into a pre-trained residual convolutional neural network and output the predicted wireless signal strength corresponding to the target location information; wherein the residual convolutional neural network includes at least a residual processing subnetwork, the residual processing subnetwork includes at least a subconvolutional layer, and the subconvolutional layer includes two sequentially connected sub-fully connected layers; The wireless signal strength distribution information determination module is configured to determine the signal strength distribution information of the target area based on the predicted wireless signal strengths corresponding to all the target location information.

8. An electronic device, characterized in that: The electronic device comprises: one or more processors; a storage device for storing one or more programs, When one or more programs are executed by one or more processors, the one or more processors implement the method for determining three-dimensional wireless signal strength according to any one of claims 1 to 6.

9. A storage medium containing computer-executable instructions, characterized in that: When executed by a computer processor, the computer executable instructions are used to perform the method for determining three-dimensional wireless signal strength according to any one of claims 1 to 6.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the computer program implements the method for determining the three-dimensional wireless signal strength according to any one of claims 1 to 6.

Images