Internet of Things Smart Phone Indoor Positioning Method and System

Abstract

The present application provides an indoor positioning method and system for an IoT smartphone. The method is applied to an indoor positioning system based on the IoT. In this solution, in response to a user's positioning operation, a control terminal sends a positioning request to a positioning server. The positioning server sends data acquisition requests to multiple access devices based on the positioning request. Each access device acquires the communication data of the mobile phone and returns the communication data to the positioning server. The communication data includes signal strength at multiple moments. The positioning server uses a position prediction model to predict the position of the user to be positioned based on the communication data sent by multiple access device stations, obtains the target position of the user to be positioned, and returns the target position to the control terminal. In this solution, artificial intelligence is used to learn the signal changes of the target during its movement in a complex indoor environment, so as to achieve accurate positioning of the user or the target to be positioned and improve the reliability of the predicted position.

Description

Technical Field

[0001] The present application relates to the field of Internet of Things and indoor positioning technology, and in particular to a method and system for indoor positioning of an Internet of Things smartphone. Background Art

[0002] As lifestyles become increasingly diverse, the number of services and activities that indoor venues can provide is also increasing. More and more users choose to carry out various activities in large indoor venues. However, the internal structure and environment of large indoor venues are complex, which makes it easy for users to get lost during activities. The demand for precise indoor positioning and navigation is increasing.

[0003] In order to solve the above problems, the technical solutions currently adopted are mainly to detect the communication signal between the user's mobile phone and the base station, measure the time required for the signal to be sent and received, calculate the distance between the mobile phone and the base station, and locate the user's position indoors by measuring the distance between multiple base stations.

[0004] However, the complex architectural structure of indoor scenes, the many wall obstructions, and the movement of users themselves will lead to unstable signals and inaccurate signal detection, making it impossible to accurately locate users indoors. Summary of the Invention

[0005] The present application provides an indoor positioning method and system for an IoT smartphone, which is used to solve the technical problems that the complex architectural structure of indoor scenes, the large number of wall obstructions, and the movement of the user themselves will lead to unstable signals and inaccurate signal detection, making it impossible to achieve accurate positioning of the user indoors.

[0006] In a first aspect, the present application provides an indoor positioning method for an IoT smartphone, which is applied to an IoT-based indoor positioning system. The indoor positioning system includes a plurality of access devices distributed in a target location, a positioning server, and a control terminal. The method includes:

[0007] In response to the user's positioning operation, the control terminal sends a positioning request to the positioning server, wherein the positioning request includes an identifier of the mobile phone of the user to be positioned;

[0008] The positioning server receives a positioning request from the control terminal and sends data acquisition requests to the multiple access devices respectively according to the positioning request, wherein the data acquisition requests include the identifier of the mobile phone;

[0009] After receiving the data acquisition request, each access device acquires the communication data of the mobile phone and returns the communication data to the positioning server, wherein the communication data includes the signal strength of the mobile phone detected at the current moment and the signal strength at multiple moments before the current moment;

[0010] The positioning server predicts the location of the user to be located using a location prediction model based on communication data sent by multiple access devices to obtain the target location of the mobile phone. The location prediction model is a model obtained by training a recursive neural network model based on a signal fingerprint database. The signal fingerprint database includes: signal strength and acquisition time of each location of the test terminal in the indoor scene detected by multiple access devices set in the indoor scene within a historical time period;

[0011] The positioning server returns the target location to the control terminal.

[0012] In a specific embodiment, the method further comprises:

[0013] The positioning server receives a positioning model configuration message sent by the cloud server, where the positioning model configuration message includes the position prediction model;

[0014] The positioning server configures the position prediction model in a positioning program based on the positioning model configuration message.

[0015] In a specific embodiment, the method further comprises:

[0016] In the positioning server, in response to the user's operation, an initial long short-term memory network LSTM model is constructed using Keras;

[0017] Normalizing the data in the signal fingerprint database to obtain a sample set, the sample set including multiple samples, each sample including: an input signal matrix and a corresponding position, the input signal strength matrix including multiple signal strength sequences corresponding to access devices, the signal strength sequence corresponding to each access device including the signal strength corresponding to any position of the test terminal in the test scenario collected by the access device and the signal strength collected at least two times before the test terminal reaches the position;

[0018] Based on the sample set, the LSTM model is trained to obtain the location prediction model.

[0019] In a specific implementation, the performing model training on the LSTM model based on the sample set to obtain the location prediction model includes:

[0020] Based on the samples in the sample set, the parameters of the memory neural network of the LSTM model are optimized by using an adaptive particle swarm algorithm to obtain an LSTM model with optimized parameters;

[0021] The LSTM model after parameter optimization is trained based on the sample set until the preset loss function converges to obtain the position prediction model.

[0022] In a specific embodiment, the method of optimizing the parameters of the memory neural network of the LSTM model by an adaptive particle swarm algorithm based on the samples in the sample set to obtain the parameter-optimized LSTM model includes:

[0023] Initializing a particle swarm of the adaptive particle swarm algorithm to obtain an initial particle swarm, wherein the initial particle swarm includes a plurality of particles, each particle being a parameter combination of the LSTM model;

[0024] Determine the objective function of the LSTM model as the fitness function of the adaptive particle swarm algorithm, and the objective function is an accuracy evaluation function;

[0025] Configuring the parameter combination corresponding to each particle in the LSTM model, inputting samples in the sample set into the LSTM model for prediction, and calculating the fitness of each particle using the fitness function based on the predicted value and actual position of the sample to obtain the fitness value of each particle;

[0026] The speed and position of each particle are updated according to the fitness value of each particle, and the fitness of each updated particle is recalculated until the number of iterations is reached to obtain the target particle swarm;

[0027] The parameter combination corresponding to the particle with the largest fitness value in the target particle swarm is configured as the optimal parameter combination into the LSTM model to complete parameter optimization.

[0028] In a specific embodiment, the optimal parameter combination includes the selected optimal number of units, number of Dense layer units, learning rate and dropout probability.

[0029] In a specific implementation, before the positioning server returns the target location to the control terminal, the method further includes:

[0030] The positioning server optimizes the target position using a filtering algorithm to obtain an optimized new target position;

[0031] Accordingly, the positioning server returns the target location to the control terminal, including:

[0032] The positioning server returns the new target location to the control terminal.

[0033] In a specific embodiment, the method further comprises:

[0034] The positioning server generates the trajectory route and movement direction of the mobile phone in real time based on the obtained target positions of the mobile phone at multiple times;

[0035] The positioning server returns the trajectory route and the moving direction to the control terminal.

[0036] In a specific embodiment, the method further comprises:

[0037] After receiving the trajectory route and the moving direction, the control terminal displays the trajectory route and the moving direction in real time on a graphical user interface, and marks the current target position of the mobile phone on the trajectory route.

[0038] A second aspect of the present application provides an indoor positioning system, comprising: a plurality of access devices distributed in a target location, a positioning server, and a control terminal;

[0039] The control terminal is used to send a positioning request to the positioning server in response to the user's positioning operation, wherein the positioning request includes the identifier of the mobile phone of the user to be positioned;

[0040] The positioning server is used to receive a positioning request from the control terminal and send data acquisition requests to the multiple access devices respectively according to the positioning request, wherein the data acquisition request includes the identifier of the mobile phone;

[0041] Each access device is configured to obtain communication data of the mobile phone according to the received data acquisition request and return the communication data to the positioning server, wherein the communication data includes the detected signal strength of the mobile phone at the current moment and the signal strength at multiple moments before the current moment;

[0042] The positioning server is further configured to predict the location of the user to be located using a location prediction model based on communication data sent by multiple access devices to obtain a target location of the mobile phone, wherein the location prediction model is a model obtained by training a recursive neural network model based on a signal fingerprint database, wherein the signal fingerprint database includes: signal strength and acquisition time of each location of the test terminal in the indoor scene detected by multiple access devices set in the indoor scene within a historical time period;

[0043] The positioning server is further configured to return the target location to the control terminal.

[0044] The third aspect of the present application further provides a computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are executed by a processor, they are used to implement the indoor positioning method for an IoT smartphone as described in any one of the first aspects.

[0045] The fourth aspect of the present application also provides a computer program product, including a computer program, which, when executed by a processor, implements the indoor positioning method for an IoT smartphone as described in any one of the first aspects.

[0046] The present application provides an indoor positioning method and system for an IoT smartphone. This method is applied to an indoor positioning system based on the IoT. In this solution, in response to a user's positioning operation, the control terminal sends a positioning request to the positioning server. The positioning server sends data acquisition requests to multiple access devices based on the positioning request. Each access device acquires the communication data of the mobile phone and returns the communication data to the positioning server. The communication data includes signal strength at multiple moments. The positioning server uses a position prediction model to predict the position of the user to be positioned based on the communication data sent by multiple access device stations, obtains the target position of the user to be positioned, and returns the target position to the control terminal. In this solution, artificial intelligence is used to learn the signal changes of the target during its movement in a complex indoor environment, so as to achieve accurate positioning of the user or the target to be positioned and improve the reliability of the predicted position. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0048] Figure 1 A schematic diagram of an indoor positioning system based on the Internet of Things provided in this application;

[0049] Figure 2 A flowchart of a first embodiment of the indoor positioning method for an IoT smartphone provided in an embodiment of the present application;

[0050] Figure 3 A flowchart of a second embodiment of the indoor positioning method for an IoT smartphone provided in an embodiment of the present application;

[0051] Figure 4 A flowchart of a third embodiment of the indoor positioning method for an IoT smartphone provided in an embodiment of the present application;

[0052] Figure 5 A flowchart of a fourth embodiment of the indoor positioning method for an IoT smartphone provided in an embodiment of the present application;

[0053] Figure 6 This is a schematic diagram of the structure of the first embodiment of the indoor positioning device for an IoT smartphone provided in an embodiment of the present application;

[0054] Figure 7This is a schematic diagram of the structure of the second embodiment of the indoor positioning device for an IoT smartphone provided in an embodiment of the present application;

[0055] Figure 8 This is a schematic diagram of the structure of the third embodiment of the indoor positioning device for an IoT smartphone provided in an embodiment of the present application;

[0056] Figure 9 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0057] The above drawings illustrate specific embodiments of the present application, which will be described in more detail below. These drawings and the textual description are not intended to limit the scope of the present application in any way, but rather to illustrate the concepts of the present application to those skilled in the art by reference to specific embodiments. DETAILED DESCRIPTION

[0058] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.

[0059] In the era of the Internet of Things (IoT), location-based services (LBSs) have become essential for people's work and daily lives. The Global Positioning System (GPS) and base station positioning technologies largely meet users' needs for LBSs in outdoor scenarios. However, 80% of our lives are spent indoors, and a significant amount of positioning needs arise indoors for individuals, service robots, and new IoT devices. However, indoor locations are often obscured by buildings, causing GNSS signals to rapidly attenuate or even be completely blocked, making them unable to meet navigation and positioning needs in these scenarios.

[0060] Currently, there are seven common indoor positioning methods, based on their principles: proximity detection, centroid positioning, multilateral positioning, triangulation, polar positioning, fingerprint positioning, and dead reckoning. Different indoor positioning methods use different observations to extract the information required by the algorithm. Based on positioning principles and observations, a variety of indoor positioning technologies have been developed. The most commonly used indoor positioning solution detects the communication signal between the user's mobile phone and the base station, measures the time required for the signal to be transmitted and received, calculates the distance between the mobile phone and the base station, and locates the user's position indoors by measuring the distance between the mobile phone and the base station. However, the complex architectural structure of indoor scenes, the presence of numerous walls, and the movement of users can lead to unstable and inaccurate signal detection, making it impossible to accurately locate the user indoors.

[0061] Based on this, while researching indoor positioning technology for complex large public venues such as shopping malls and airports, the inventors discovered that a method and system for indoor positioning using an IoT smartphone, based on artificial intelligence (AI), could address the limited positioning accuracy of existing technologies. This method also offers superior scalability, accuracy, cost, complexity, and response time. This solution first develops a terminal application for data collection and real-time data transmission for the IoT smartphone-based indoor positioning system. This application collects received signal strength indication (RSSI) data during daily operations and then sends this RSSI data to a positioning server. Simultaneously, real-time monitoring software is developed on the positioning server, allowing the server to communicate directly with the smartphone to receive the detected RSSI dataset. The smartphone collects RSSI data from each access device at each reference point (i.e., each location in the indoor scene, which can be pre-marked). After all the data is collected and transmitted, the positioning server processes the data to form a signal fingerprint database. The next step is to train a recursive neural network algorithm based on the data in the signal fingerprint database, ultimately generating a location prediction model. Finally, in the online positioning stage, when real-time positioning is required, the positioning server collects the real-time RSSI of the mobile phone to be located through various access devices in the target location, uses the pre-obtained location prediction model to predict the target location, and feeds it back to the control terminal or the user's mobile phone.

[0062] The technical solutions involved in this application are described in detail below through several specific embodiments.

[0063] Figure 1 A schematic diagram of an indoor positioning system based on the Internet of Things provided in this application is shown in FIG. Figure 1As shown, the indoor positioning system needs to include at least a positioning server and multiple access devices, which are respectively set up at various locations in the target place. These access devices can be Wi-Fi hotspots or wireless access points (APs) such as base stations, which can detect the strength of communication signals. This solution does not impose specific restrictions on this.

[0064] The positioning server can be a cloud server, a server dedicated to positioning, or multiple hybrid servers. The model training can be implemented separately on one of the servers and then configured on the server used for positioning. This solution does not impose any restrictions on this.

[0065] Furthermore, during the actual application of the indoor positioning system, the system may also include a control terminal. It should be understood that the control terminal may be a public service terminal in large target places such as shopping malls, or it may be a mobile phone, computer or other terminal of other associated users who need to find the user to be located. In addition, it may also be the terminal device of the user to be located himself, and this solution does not impose any restrictions on this.

[0066] Based on the above application scenarios, the implementation process of the indoor positioning method for IoT smartphones provided in this application is described in detail below.

[0067] Example 1

[0068] Figure 2 This is a flow chart of the first embodiment of the method for indoor positioning of an IoT smartphone provided in the embodiment of the present application, as shown in FIG. Figure 2 As shown, the indoor positioning method of the IoT smartphone specifically includes the following steps:

[0069] S101: In response to a positioning operation by a user, the control terminal sends a positioning request to the positioning server.

[0070] In this step, when a user needs to be located in an indoor setting such as a shopping mall, airport, or large indoor amusement park, a positioning operation can be initiated from the control terminal. The user on the control terminal operates the control terminal and enters identification information such as a mobile phone number that uniquely identifies the user's mobile phone to initiate a positioning server. In response to this operation, the control terminal can send a positioning request to the positioning server of the entire system, which includes the identification of the user's mobile phone to be located.

[0071] S102: The positioning server receives a positioning request from the control terminal, and sends data acquisition requests to multiple access devices according to the positioning request.

[0072] In this step, the data acquisition request includes the identification of the mobile phone.

[0073] For the positioning server, it receives the positioning request sent by the control terminal, which includes the identification of the mobile phone to be located. Based on this, the positioning server sends a data acquisition request carrying the identification of the mobile phone to multiple access devices in the current indoor positioning system, so that these access devices can obtain communication data with the mobile phone based on the identification.

[0074] S103: After receiving the data acquisition request, each access device acquires the communication data of the mobile phone.

[0075] S104: Returning communication data to the positioning server, the communication data including the detected signal strength of the mobile phone at the current moment and the signal strength at multiple moments before the current moment.

[0076] In the above two steps, for each access device, after receiving the data acquisition request, the communication data is obtained based on the historical communication data and the ongoing communication situation. It should be understood that the communication data in this solution generally requires the communication signal strength at multiple times. For each access device, it can initiate communication with the mobile phone after receiving the data acquisition request, and detect the signal strength at multiple times to obtain the communication data, or it can detect the signal strength at the current moment and obtain the signal strength of several previous moments from the historical communication data to obtain the communication data. This solution does not impose any restrictions on this.

[0077] After acquiring the communication data, each access device returns the acquired communication data to the positioning server for analysis and processing.

[0078] In this solution, it should be understood that due to the complex architectural structures of large indoor venues, even if a large number of access devices are deployed and the user to be located is indeed carrying the mobile phone within the large venue, some access devices may not detect the mobile phone at all. In this case, the communication data obtained by the access device may be 0 or empty. For example, if the configuration is to determine location information based on data at three times, the communication data returned by the access device that did not detect the mobile phone may include the signal strength corresponding to the three times being 0 or empty.

[0079] S105: The positioning server predicts the location of the user to be positioned using a location prediction model based on the communication data sent by the multiple access devices, and obtains the target location of the mobile phone.

[0080] In this step, after the positioning server obtains the communication data collected for the mobile phone from each access device, it can analyze and process this data using a pre-acquired location prediction model to determine the mobile phone's target location at the next moment. In this solution, the location prediction model is a recursive neural network model trained based on a signal fingerprint database. The signal fingerprint database includes the signal strength and acquisition time of each location of the test terminal in the indoor scene, as detected by multiple access devices set up in the scene over a historical time period.

[0081] In the specific implementation, the LSTM deep learning model can be used as the neural recurrent network model for position prediction because it can effectively memorize and process information from time series data. Other machine learning models can also be used, and this solution does not impose any restrictions on this.

[0082] S106: The positioning server returns the target location to the control terminal.

[0083] After the positioning server determines the target of the mobile phone, it can return the target position to the control terminal, so that the control terminal can display the target position, so that the user in need can know the specific position of the user to be located.

[0084] In this solution, a fingerprint database is established in advance by collecting signal fingerprints at various locations in large indoor scenes, a location prediction model is constructed based on a recursive neural network model, and the model is trained based on the fingerprint database to obtain the location prediction model. Finally, the model is applied to some indoor positioning scenarios. When searching for the elderly, children, or for other positioning needs, the positioning prediction model has learned the signal instability factors caused by building structures and other factors at various locations. That is, artificial intelligence is used to learn the signal changes of the target during its movement in a complex indoor environment, so as to achieve accurate positioning of the user or the target to be located, effectively improving the reliability of positioning prediction.

[0085] Based on the above embodiments, it can be seen that in order to achieve accurate positioning in a certain indoor scene, it is necessary to collect some data in this scene in advance to build a signal fingerprint database for model training, and then use the model in the actual positioning scene. The positioning server configures the location prediction model including at least the following two implementation solutions:

[0086] The first implementation method is to train the model on other servers, and then configure the obtained location prediction model to be used in the positioning server.

[0087] For example, after the data in the signal fingerprint database is standardized, the location prediction model can be trained on the cloud server. After training, the location prediction model is sent to the positioning server. The positioning server receives the positioning model configuration message sent by the cloud server, and the positioning model configuration message includes the location prediction model. Then, based on the positioning model configuration message, the location prediction model is configured in the positioning program for subsequent location prediction applications.

[0088] The second implementation method is to use the positioning server that provides positioning services for the indoor scene to collect data and train models to obtain a location prediction model, which is finally applied in the positioning service.

[0089] Specifically, by setting up a test terminal, tests can be performed at each location in the entire indoor scene. The specific test method can be that the test user carries the test device to conduct multiple tests along different direction paths, detects the signal strength and collection time and stores them in the fingerprint database. After the complete data is collected, it is standardized in the positioning server to obtain a sample set, and then the model is trained to obtain the location prediction server.

[0090] Below, taking the selection of the LSTM model and the implementation of model training in the positioning server as an example (the process of model training on other servers is similar), the specific training process of the location prediction model provided in this application is illustrated.

[0091] Before describing the model training process in this embodiment, we will discuss the construction of the signal fingerprint database. First, location reference points are set within the test site (i.e., the indoor scene where the final model will be applied). This means that all areas in the entire environment are divided into simple locations. Especially for environments with complex structures and layouts, where signals are affected by building structures and external devices, location reference points need to be set at smaller intervals within these areas.

[0092] After the location reference point is set, the test user can wear a terminal that collects signal strength (for example, a smart phone, a phone watch, etc.), and the access device can collect signal strength at a preset frequency (the frequency can be set according to actual conditions, such as collecting once every one second), taking into account both static and mobile scenarios. For static scenarios, the test user holds the test terminal to collect signal strength at each test location. For mobile scenarios, the user carries the test terminal and walks along a pre-designed route while recording the signal strength. For example, there are 6 access devices in the test scenario, and their location coordinates are (3,1), (10,1), (19,1), (3,6), (10,6), and (19,6). For each location, the set of collected signal strengths (which can be WIFI signals, cellular network signals, or the strength of signals from other types of wireless communication technologies) can be defined as ,in Indicates the The signal strength of the access device is If an access device cannot be detected at the current location, It is 0 or empty, or -100 (usually the default minimum value of RSSI). Assume that the signal strength set is recorded at a certain location As the user moves, the test device moves to another location to detect the If the first access device cannot be detected, the The signal strength detected by each access device can be empty, 0 or the lowest value of the signal strength, but there are other access devices j that detect the signal strength and record the signal strength, and the signal strength set of the location is obtained. .

[0093] Optionally, since the walking speeds of different users and the same user at different times are not consistent, in order to obtain the true position corresponding to the signal strength detected on the moving path, the user is allowed to maintain a constant speed within the same path as much as possible, and then the user's walking speed is calculated by dividing the path length by the total time. Then, the timestamps of the collected signal strengths are interpolated between the starting point and the end point of the preset route to obtain the true position of each sampled signal strength on the moving path.

[0094] After collecting all the data using the above method, each area of the test scene can be divided into grids of equal size (for example, 1m*1m). The size of the grid can be modified. At each grid vertex, the signal strength of all connected devices in the test scene is collected, and the collected signal strength and location data are entered into the signal fingerprint database. For example, the received signal strength (RSSI) of a Wi-Fi signal is determined by the location of the receiving device. As the communication distance increases, the signal strength changes due to environmental influences during propagation, and the degree of signal strength attenuation is affected by distance. However, through the signal strength collection method described above, the signal strength changes at each location are recorded in the database for learning by the machine learning model.

[0095] Example 2

[0096] Figure 3 This is a flow chart of the second embodiment of the method for indoor positioning of an IoT smartphone provided in the embodiment of the present application, as shown in FIG. Figure 3 As shown in the figure, the process of model training in the positioning server is as follows:

[0097] S201: In the positioning server, in response to the user's operation, an initial LSTM model is constructed using Keras.

[0098] In this step, we use the LSTM model built using Keras. This LSTM model can effectively memorize and process information from time series data. The main structure consists of two layers: an LSTM layer and a fully connected layer.

[0099] S202: Standardize the data in the signal fingerprint database to obtain a sample set, where the sample set includes multiple samples, each sample includes: an input signal matrix and a corresponding position, the input signal strength matrix includes multiple signal strength sequences corresponding to access devices, and the signal strength sequence corresponding to each access device includes the signal strength corresponding to the test terminal at any position in the test scenario collected by the access device and the signal strength collected at at least two moments before the test terminal reaches the said position.

[0100] Before training the model, the data in the pre-collected fingerprint database needs to be processed to conform to the actual model input and output data structure. Because the signal strength and time previously collected by the test terminal in the test scenario are discrete, they need to be organized into samples according to certain rules. Specifically, the final location prediction model requires the signal strength of multiple access devices at several consecutive moments as input to predict the target's location at the next moment. Therefore, the data in the signal fingerprint database also needs to be standardized according to this rule to obtain training samples that conform to the model's input and output structure.

[0101] In this embodiment, it should also be understood that, because data collected by multiple access devices must be comprehensively considered in actual implementation, the LSTM model input must be designed as different input units. Different input units are used to input sequences collected by different access devices. The length of each sequence represents the number of previous signal strength moments selected for predicting the current location, for example, 3 or 5 (the specific length can be improved based on actual needs). The data in the signal fingerprint database is recombined to obtain a signal matrix input to the LSTM model and multiple samples consisting of the positions corresponding to the signal matrix. For each input signal matrix, the signal strength sequence corresponding to multiple access devices (e.g., the aforementioned six access devices) is included. The signal strength sequence corresponding to each access device includes the signal strength of the test terminal at any location in the test scenario, collected by the access device, as well as the signal strength at several moments before the test terminal reached the location. Through this processing method, a sample set consisting of multiple samples can be obtained.

[0102] S203: Based on the sample set, the LSTM model is trained to obtain a location prediction model.

[0103] In this step, after processing the sample set, the LSTM model can be trained based on the samples in the sample set. In each training iteration, data is fed into the model, and the loss value is calculated according to the preset loss function. The model weights are then updated through the optimizer. The updated model is evaluated and the model parameters are adjusted and optimized based on the evaluation results. The optimization also includes optimizing the training method and model architecture until the evaluation meets the requirements and the location prediction model is obtained.

[0104] In one possible implementation, step 203 of the above embodiment can be specifically implemented by optimizing the parameters of the LSTM model's neural network using an adaptive particle swarm algorithm based on the samples in the sample set, thereby obtaining a parameter-optimized LSTM model. The parameter-optimized LSTM model is then trained based on the sample set until a preset loss function converges, thereby obtaining the position prediction model.

[0105] Specifically, the specific implementation steps for optimizing the parameters of the LSTM model's memory neural network using the adaptive particle swarm algorithm are as follows:

[0106] Step 1: Initialize the particle swarm of the adaptive particle swarm algorithm to obtain an initial particle swarm, wherein the initial particle swarm includes a plurality of particles, each particle being a parameter combination of the LSTM model.

[0107] Step 2: Determine the objective function of the LSTM model as the fitness function of the adaptive particle swarm algorithm, and the objective function is an accuracy evaluation function.

[0108] Step 3: Configure the parameter combination corresponding to each particle in the LSTM model, input the samples in the sample set into the LSTM model for prediction, and calculate the fitness of each particle using the fitness function based on the predicted value and actual position of the sample to obtain the fitness value of each particle.

[0109] Step 4: update the speed and position of each particle according to the fitness value of each particle, and recalculate the fitness of each updated particle until the number of iterations is reached to obtain the target particle swarm;

[0110] Step 5: The parameter combination corresponding to the particle with the largest fitness value in the target particle swarm is configured as the optimal parameter combination into the LSTM model to complete parameter optimization.

[0111] Optionally, the optimal parameter combination includes a selected optimal number of units, number of Dense layer units, learning rate, and dropout probability.

[0112] Before introducing the technology of this embodiment in detail, the basic steps of the PSO algorithm are first introduced. First, the number of particles is set, the positions and velocities of all particles are initialized, the optimal solution of the individual is set as the current position, and the optimal individual in the group is used as the current group optimal solution; each iteration calculates the fitness function value of each particle, if the current fitness function value is better than the individual historical optimal value, the individual historical optimal value is updated; if the current fitness function value is better than the global historical optimal value, the global historical optimal value is updated; when the number of iterations reaches the maximum, the optimization process stops.

[0113] This example improves upon the traditional PSO algorithm by introducing a chaotic sequence initialization to replace the random initialization of particle swarm positions in PSO. The pseudo-random initialization in standard PSO can cause the swarm to cluster in a localized area, impacting overall search performance. A chaotic sequence typically exhibits better randomness, distributing the initial swarm positions more widely across the search space, thereby increasing the chances of finding the global optimal solution.

[0114] In the specific implementation, the Tent chaotic map is used to initialize the particle population position. The formula is as follows:

[0115] ,

[0116] in, ;

[0117] In the above formula, is a chaotic sequence in the interval [0,1]; is the chaos value, set to 0.7; is the population number; is the number of iterations; is a chaotic sequence that conforms to the particle range; and are the upper and lower boundaries of the particle range.

[0118] In addition, in the process of adjusting particle weights, this scheme adopts a nonlinear weight adjustment strategy. Compared with the linear weight adjustment, the nonlinear decreasing weight adjustment strategy is adopted to make the particle weights In the early stage of the search, the particle's global search ability is stronger; In the later stages of the search, the particle's local optimization ability is stronger, which increases the chance of finding the optimal solution. The improved inertia weight expression is:

[0119]

[0120] Where, and are the maximum inertia weight and the minimum inertia weight, which are set to 0.7 and 0.4 respectively; is the maximum number of iterations.

[0121] Furthermore, the learning factor of the particle 、 At the same time, optimization is performed to Decrease while increasing In the early stage of the search, the self-learning ability of particles is emphasized, while in the later stage, the social learning ability of particles is emphasized to increase the chance of obtaining the global optimal solution. The improved learning factor expression is:

[0122] ;

[0123] ;

[0124] In the above formula, For the generation, and is the maximum value of the individual and group learning factors, both set to 2.5; and is the minimum value of the individual and group learning factors, both set to 1.

[0125] Based on the above improvements to the PSO algorithm, the adaptive particle swarm algorithm used in this scheme is obtained.

[0126] After completing the algorithm improvement, based on the above-mentioned improved adaptive particle swarm algorithm and the samples in the sample set, the parameters such as the number of hidden layer units and the learning rate in the LSTM model were tuned to obtain the optimal parameter combination consisting of the optimal number of units, the number of dense layer units, the learning rate, and the dropout probability. The specific process is as follows:

[0127] The particle swarm is initialized by using chaotic sequence initialization to generate a certain number of particles to form a The initial population of particles is generated, and each particle represents a parameter combination in the LSTM model. The parameter combination may include at least one parameter that needs to be optimized. Preferably, the four parameters mentioned above, namely the number of units, the number of Dense layer units, the learning rate, and the dropout probability, can be selected.

[0128] The root mean square error between the LSTM model's predictions and the actual values can be used as a loss function to adapt the fitness of the particle swarm algorithm. The parameter combination corresponding to each particle can be configured in the LSTM model. Samples from the sample set are then input into the LSTM model. The fitness of each particle is calculated according to the fitness function to obtain a fitness value for each particle. The speed and position of each particle are updated based on the fitness value, and the fitness of each updated particle is recalculated until the number of iterations is reached, resulting in the globally optimal target particle. The parameter combination corresponding to the target particle's position can be configured in the LSTM model.

[0129] Then, the other parameters and activation functions of the optimized LSTM model are configured, and the calculation function of the model training is set. The sample set is divided into a training set and a validation set. The samples in the training set are used to train the optimized LSTM model, and the trained model is tested and verified based on the samples in the validation set, and error analysis is performed until the position prediction model in this application is obtained.

[0130] In the technical solution provided in this embodiment, the improved adaptive particle swarm algorithm optimizes the parameters of the LSTM model, which can improve the prediction ability of the LSTM model, that is, improve the accuracy of position prediction.

[0131] In one specific implementation of this solution, the LSTM model is selected for tracking during training. The vanishing gradient problem can make it difficult to model long-term dependencies. Therefore, to speed up convergence, the ReLU function can be selected as the activation function. The ReLU function's derivative is always 1 when the input is greater than 0, eliminating the vanishing gradient problem. This effectively solves the problem. This makes it easier for the LSTM model to learn and understand data features during training, resulting in better convergence and higher prediction accuracy.

[0132] Example 3:

[0133] Figure 4 This is a flow chart of the third embodiment of the method for indoor positioning of an IoT smartphone provided in the embodiment of the present application, as shown in FIG. Figure 4 As shown, based on any of the above embodiments, in another specific implementation of the indoor positioning method for an IoT smartphone, before the positioning server returns the target position to the control terminal, it also includes:

[0134] S107: The positioning server optimizes the target position using a filtering algorithm to obtain an optimized new target position.

[0135] Accordingly, S106 can be specifically implemented as follows:

[0136] S1061: The positioning server returns the new target location to the control terminal.

[0137] In the technical solution of this embodiment, in order to further improve the accuracy of the predicted position, filtering can be integrated with the machine learning in the aforementioned solution during the prediction processing in the positioning server, and the predicted target position can be corrected by the filtering algorithm, which can reduce the error caused by the sudden fluctuation of signal strength caused by the complex indoor environment, especially for people who are always moving. In this application, a position prediction model is used to first perform predictive positioning, and then the predicted target position is optimized as the input of the filter.

[0138] In this solution, because the motion states of the users to be located are continuous, correlated, and have certain regularities, the filtering method can provide better positioning accuracy and positioning stability. The particle filter algorithm is an approximate estimation method that uses particle sampling simulation to simulate the state space at any time. The particle filter updates the weights and states of the particles based on information and current observations (retaining the particles with high weights and discarding the particles with low weights, and then adding up the weighted predicted positions of the particles that remain, which is the current moment). Finally, the position trajectory can be output under the premise of high precision. Therefore, this solution preferably uses particle filtering to optimize the target position.

[0139] Specifically, the algorithm process of optimizing the target position using the particle filter algorithm is as follows:

[0140] (1) In the entire area of the preset range around the predicted target position, if the total number of position reference points in the preset range is N, then these position reference points are taken as the total number of The particles are randomly distributed and the state of the particle swarm is initialized, including the position of each particle , motion direction angle and movement speed , and initialize the weight value of each particle to .in and Obeying uniform distribution, is a constant, but the speed of the user to be located will not be constant when moving, and there will be speed noise expressed as , and obeys the standard normal distribution.

[0141] (2) The particles move according to the particle state equation below, and the next position of the particle group is predicted (( , ) indicates that the particle is Position at the moment):

[0142]

[0143] in, For particles in The direction of motion at the moment.

[0144] (3) Update the particle weights based on the probability information that the Euclidean distance between the predicted value and the true position of each particle satisfies the Gaussian distribution function;

[0145] (4) Calculate the number of effective particles. If the number of effective particles is less than a certain threshold, resample the particles and normalize the weights of the resampled particles.

[0146] (5) The positions represented by the particles resampled in (4) are weighted and accumulated, and the weighted sum obtained is the new target position after optimization.

[0147] As the positioning target moves, in the subsequent position prediction process, the aforementioned particle filter optimization process is repeated each time the target is predicted to be a child. This not only improves the positioning accuracy, but also enables real-time position tracking.

[0148] The IoT smartphone positioning method provided in this embodiment introduces a filtering algorithm to further optimize the target location predicted by the machine learning method, thereby improving positioning accuracy. In this preferred solution, a particle filtering algorithm is selected for fusion. During the optimization process, the Euclidean distance between the true and estimated positions is used as the position error, and the time complexity (runtime) is used as the performance metric for evaluating the positioning algorithm. This effectively reduces errors caused by sudden fluctuations in signal strength due to indoor obstacles and multipath, improving positioning accuracy while also achieving rapid processing speed and excellent effectiveness.

[0149] Example 4:

[0150] Figure 5This is a flow chart of the fourth embodiment of the method for indoor positioning of an IoT smartphone provided in the embodiment of the present application, as shown in FIG. Figure 5 As shown, based on any of the above embodiments, the indoor positioning method of the Internet of Things smartphone further includes the following steps:

[0151] S301: The positioning server generates the trajectory route and movement direction of the mobile phone in real time based on the obtained target positions of the mobile phone at multiple times.

[0152] S302: The positioning server returns the trajectory route and moving direction to the control terminal.

[0153] S303: After receiving the trajectory route and the moving direction, the control terminal displays the trajectory route and the moving direction in real time on the graphical user interface, and marks the current target position of the mobile phone on the trajectory route.

[0154] In this solution, since the positioning server can obtain the location of the mobile phone in real time, after accumulating the location points at multiple times, it can generate the movement trajectory of the mobile phone according to these trajectory points, that is, the trajectory route of the user to be located. Based on the target positions at several times, the movement direction of the user to be located can also be determined. After obtaining the trajectory route and movement direction, the trajectory route and movement direction are returned to the control terminal, which can be displayed on the graphical user interface and at the same time mark the current target position of the mobile phone of the user to be located on the route.

[0155] In this way, the solution can intuitively provide the control terminal with the movement direction, route and location of the user to be located, making it easier for the user on the demand side to find the target and improving the efficiency and accuracy of target finding.

[0156] Furthermore, this implementation approach can be integrated with the interfaces of large shopping malls, security, medical, scenic spots, and other institutional platforms. By providing a three-dimensional visual representation of one's own or other party's location and target orientation, visitors can accurately locate their location and receive real-time feedback on their devices. This is crucial for autonomous robot positioning and finding lost users. Furthermore, this solution can collect a wealth of useful information before, during, and after an event, providing timely warnings of potential risk events.

[0157] Furthermore, the display in the control terminal can also choose to display three-dimensional space. Especially for scene spaces with complex structures, the entire indoor scene can be pre-modeled in three dimensions through image acquisition, combined with synchronous positioning and map construction technology to obtain a three-dimensional space model. After obtaining the target position, trajectory route and movement direction of the user to be located through the aforementioned technical solution, a virtual object can be created in the three-dimensional space model, and the virtual object, trajectory route and movement direction can be superimposed in the form of a virtual image to the corresponding position in the three-dimensional space model and displayed on the interface. The specific trajectory and direction display can include dynamic arrows, path lines, etc. These elements can be updated in real time as the target moves. In this way, the intuitiveness of the interaction can be improved and the efficiency can be improved when searching for lost people.

[0158] In summary, this application has developed a method and system for indoor positioning of IoT smartphones. This method combines signal strength positioning technology, signal strength fingerprint feature extraction, and intelligent algorithms, significantly improving the accuracy and efficiency of location prediction for personnel in large shopping malls. By collecting key signal fingerprint information through an automated system, preferably, an IPSO-LSTM combined with a filter fusion positioning algorithm is used to achieve mobile phone positioning, effectively improving the accuracy of indoor mobile phone positioning prediction.

[0159] In the technical solution of the present application, it should be understood that the aforementioned method embodiments are only exemplified by taking smartphones as the target. In actual applications, any terminal capable of wireless communication can adopt the above technical solution to achieve accurate positioning prediction, such as phone watches, tablet computers, smart bracelets, anti-lost wearable devices, etc. This solution does not impose any restrictions on this.

[0160] Figure 6 This is a structural diagram of the first embodiment of the indoor positioning device for an IoT smartphone provided in the embodiment of the present application, as shown in FIG. Figure 6 As shown, the indoor positioning device for an IoT smartphone is used to implement the method for positioning the server side in the aforementioned method embodiment. Specifically, the indoor positioning device 10 for an IoT smartphone includes:

[0161] The receiving module 11 is configured to receive a positioning request from the control terminal;

[0162] A sending module 12 is configured to send data acquisition requests to the multiple access devices respectively according to the positioning request, wherein the data acquisition requests include the identifier of the mobile phone;

[0163] The receiving module 11 is further configured to receive communication data returned by each access device, wherein the communication data includes the detected signal strength of the mobile phone at the current moment and the signal strength at multiple moments before the current moment.

[0164] a processing module 13 configured to predict the location of the user to be located using a location prediction model based on communication data sent by multiple access devices to obtain a target location of the mobile phone, wherein the location prediction model is a model obtained by training a recursive neural network model based on a signal fingerprint database, wherein the signal fingerprint database includes: signal strength and acquisition time at each location of the test terminal in the indoor scene, detected by multiple access devices set in the indoor scene within a historical time period;

[0165] The sending module 12 is further configured to return the target location to the control terminal.

[0166] Optionally, the receiving module 11 is further configured to receive a positioning model configuration message sent by a cloud server, wherein the positioning model configuration message includes the position prediction model;

[0167] The processing module 13 is further configured to configure the position prediction model in a positioning program based on the positioning model configuration message.

[0168] Optionally, the processing module 13 is further configured to:

[0169] In the positioning server, in response to the user's operation, an initial long short-term memory network LSTM model is constructed using Keras;

[0170] Normalizing the data in the signal fingerprint database to obtain a sample set, the sample set including multiple samples, each sample including: an input signal matrix and a corresponding position, the input signal strength matrix including multiple signal strength sequences corresponding to access devices, the signal strength sequence corresponding to each access device including the signal strength corresponding to any position of the test terminal in the test scenario collected by the access device and the signal strength collected at least two times before the test terminal reaches the position;

[0171] Based on the sample set, the LSTM model is trained to obtain the location prediction model.

[0172] Optionally, the processing module is specifically configured to:

[0173] Based on the samples in the sample set, the parameters of the memory neural network of the LSTM model are optimized by using an adaptive particle swarm algorithm to obtain an LSTM model with optimized parameters;

[0174] The LSTM model after parameter optimization is trained based on the sample set until the preset loss function converges to obtain the position prediction model.

[0175] Optionally, the processing module 13 is further specifically configured to:

[0176] Initializing a particle swarm of the adaptive particle swarm algorithm to obtain an initial particle swarm, wherein the initial particle swarm includes a plurality of particles, each particle being a parameter combination of the LSTM model;

[0177] Determine the objective function of the LSTM model as the fitness function of the adaptive particle swarm algorithm, and the objective function is an accuracy evaluation function;

[0178] Configuring the parameter combination corresponding to each particle in the LSTM model, inputting samples in the sample set into the LSTM model for prediction, and calculating the fitness of each particle using the fitness function based on the predicted value and actual position of the sample to obtain the fitness value of each particle;

[0179] The speed and position of each particle are updated according to the fitness value of each particle, and the fitness of each updated particle is recalculated until the number of iterations is reached to obtain the target particle swarm;

[0180] The parameter combination corresponding to the particle with the largest fitness value in the target particle swarm is configured as the optimal parameter combination into the LSTM model to complete parameter optimization.

[0181] Optionally, the optimal parameter combination includes a selected optimal number of units, number of Dense layer units, learning rate and dropout probability.

[0182] Optionally, the processing module 13 is further configured to: optimize the target position using a filtering algorithm to obtain an optimized new target position.

[0183] Correspondingly, the sending module 12 is specifically configured to return the new target location to the control terminal.

[0184] Optionally, the processing module 13 is further configured to generate a trajectory route and a moving direction of the mobile phone in real time based on the acquired target positions of the mobile phone at multiple moments;

[0185] The sending module 12 is further specifically configured to return the trajectory route and the moving direction to the control terminal.

[0186] Figure 7 This is a structural diagram of the second embodiment of the indoor positioning device for an IoT smartphone provided in the embodiment of the present application, as shown in FIG. Figure 7 As shown, the IoT smartphone indoor positioning device is used to implement any access device side method in the aforementioned method embodiments. Specifically, the IoT smartphone indoor positioning device 20 includes:

[0187] The receiving module 21 is configured to receive a data acquisition request sent by a positioning server, wherein the data acquisition request includes an identifier of the mobile phone;

[0188] a processing module 22 configured to obtain communication data of the mobile phone according to the data acquisition request, wherein the communication data includes the detected signal strength of the mobile phone at a current moment and the signal strength at multiple moments before the current moment;

[0189] The sending module 23 returns the communication data to the positioning server.

[0190] Figure 8 This is a structural diagram of the third embodiment of the indoor positioning device for an IoT smartphone provided in the embodiment of the present application, as shown in FIG. Figure 8 As shown, the indoor positioning device for an IoT smartphone is used to implement any of the control terminal side methods in the aforementioned method embodiments. Specifically, the indoor positioning device 30 for an IoT smartphone includes:

[0191] The sending module 33 is configured to send a positioning request to the positioning server in response to the user's positioning operation, wherein the positioning request includes an identifier of the mobile phone of the user to be positioned;

[0192] Receiving module 31, used for receiving the target location returned by the positioning server;

[0193] The display module 32 is configured to display the target location on a graphical user interface.

[0194] Optionally, the receiving module 31 is further configured to receive the trajectory route and movement direction sent by the positioning server;

[0195] The display module 32 is further configured to display the trajectory route and the moving direction in real time on a graphical user interface, and mark the current target location of the mobile phone on the trajectory route.

[0196] The IoT smartphone indoor positioning device provided in any of the above embodiments is used to execute the technical solution in the above method embodiments. Its implementation principles and technical effects are similar and will not be repeated here.

[0197] Figure 9 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application, such as Figure 9 As shown, the electronic device 400 includes: a processor 401, a memory 402 and a communication interface 403;

[0198] The memory 402 is used to store computer programs, and the processor 401 is used to execute the computer programs to implement the technical solution in any of the aforementioned method embodiments.

[0199] In actual implementation, the electronic device 400 can be implemented as any one of a positioning server, a control terminal, and an access device, and this solution does not impose any limitation on this.

[0200] This embodiment provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the microscopic measurement data processing method in the above embodiment.

[0201] This embodiment further provides a computer program product, including a computer program. When the computer program is executed by a processor, the microscopic measurement data processing method provided by any one of the above embodiments is implemented.

[0202] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present application that follow the general principles of the present application and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0203] It should be understood that the present application is not limited to the exact structure described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A method for indoor positioning of an Internet of Things smartphone, characterized in that: Applied to an indoor positioning system based on the Internet of Things, the indoor positioning system includes multiple access devices distributed in a target location, a positioning server, and a control terminal. The method includes: In response to the user's positioning operation, the control terminal sends a positioning request to the positioning server, wherein the positioning request includes an identifier of the mobile phone of the user to be positioned; The positioning server receives a positioning request from the control terminal and sends data acquisition requests to the multiple access devices respectively according to the positioning request, wherein the data acquisition requests include the identifier of the mobile phone; After receiving the data acquisition request, each access device acquires the communication data of the mobile phone and returns the communication data to the positioning server, wherein the communication data includes the signal strength of the mobile phone detected at the current moment and the signal strength at multiple moments before the current moment; The positioning server uses a location prediction model to predict the location of the user to be located based on the communication data sent by multiple access devices to obtain the target location of the mobile phone. The location prediction model is based on a signal fingerprint database, and is trained using an adaptive particle swarm algorithm to optimize the parameters of an LSTM model. The signal fingerprint database includes: the signal strength and acquisition time of each location of the test terminal in the indoor scene detected by multiple access devices set in the indoor scene within a historical time period; The positioning server returns the target location to the control terminal; Among them, the adaptive particle swarm algorithm is used to optimize the parameters of the LSTM model, including: Initializing a particle swarm of the adaptive particle swarm algorithm to obtain an initial particle swarm, wherein the initial particle swarm includes a plurality of particles, each particle being a parameter combination of the LSTM model; Determine the objective function of the LSTM model as the fitness function of the adaptive particle swarm algorithm, and the objective function is an accuracy evaluation function; Configuring the parameter combination corresponding to each particle in the LSTM model, inputting samples in a sample set into the LSTM model for prediction, and calculating the fitness of each particle using the fitness function based on the predicted value and actual position of the sample to obtain the fitness value of each particle, wherein the sample set is obtained by standardizing the signal fingerprint database; The speed and position of each particle are updated according to the fitness value of each particle, and the fitness of each updated particle is recalculated until the preset number of iterations is reached to obtain the target particle swarm; The parameter combination corresponding to the particle with the largest fitness value in the target particle swarm is configured as the optimal parameter combination into the LSTM model to complete parameter optimization.

2. The method for indoor positioning of an IoT smartphone according to claim 1, wherein: The method further comprises: The positioning server receives a positioning model configuration message sent by the cloud server, where the positioning model configuration message includes the position prediction model; The positioning server configures the position prediction model in a positioning program based on the positioning model configuration message.

3. The method for indoor positioning of an IoT smartphone according to claim 1, wherein: The method further comprises: In the positioning server, in response to the user's operation, an initial long short-term memory network LSTM model is constructed using Keras; Standardizing the data in the signal fingerprint database to obtain a sample set, the sample set including multiple samples, each sample including: an input signal matrix and a corresponding position, the input signal strength matrix including multiple signal strength sequences corresponding to access devices, the signal strength sequence corresponding to each access device including the signal strength corresponding to any position of the test terminal in the test scenario collected by the access device and the signal strength collected at least two times before the test terminal reaches the position; Based on the sample set, the LSTM model is trained to obtain the location prediction model.

4. The method for indoor positioning of an IoT smartphone according to claim 3, wherein: The step of training the LSTM model based on the sample set to obtain the location prediction model includes: Based on the samples in the sample set, the parameters of the neural network of the LSTM model are optimized by using an adaptive particle swarm algorithm to obtain an LSTM model with optimized parameters; The LSTM model after parameter optimization is trained based on the sample set until the preset loss function converges to obtain the position prediction model.

5. The method for indoor positioning of an Internet of Things smartphone according to claim 1, wherein: The optimal parameter combination includes the selected optimal number of units, number of Dense layer units, learning rate and dropout probability.

6. The method for indoor positioning of an IoT smartphone according to any one of claims 1 to 5, characterized in that: Before the positioning server returns the target location to the control terminal, the method further includes: The positioning server optimizes the target position using a filtering algorithm to obtain an optimized new target position; Accordingly, the positioning server returns the target location to the control terminal, including: The positioning server returns the new target location to the control terminal.

7. The method for indoor positioning of an IoT smartphone according to claim 6, wherein: The method further comprises: The positioning server generates the trajectory route and movement direction of the mobile phone in real time based on the obtained target positions of the mobile phone at multiple times; The positioning server returns the trajectory route and the moving direction to the control terminal.

8. The method for indoor positioning of an IoT smartphone according to claim 7, wherein: The method further comprises: After receiving the trajectory route and the moving direction, the control terminal displays the trajectory route and the moving direction in real time on a graphical user interface, and marks the current target position of the mobile phone on the trajectory route.

9. An indoor positioning system, characterized in that: include: Multiple access devices, positioning servers, and control terminals distributed in the target location; The control terminal is used to send a positioning request to the positioning server in response to the user's positioning operation, wherein the positioning request includes the identifier of the mobile phone of the user to be positioned; The positioning server is used to receive a positioning request from the control terminal and send data acquisition requests to the multiple access devices respectively according to the positioning request, wherein the data acquisition request includes the identifier of the mobile phone; Each access device is configured to obtain communication data of the mobile phone according to the received data acquisition request and return the communication data to the positioning server, wherein the communication data includes the detected signal strength of the mobile phone at the current moment and the signal strength at multiple moments before the current moment; The positioning server is further configured to predict the location of the user to be located using a location prediction model based on communication data sent by multiple access devices to obtain the target location of the mobile phone, wherein the location prediction model is based on a signal fingerprint database and is trained using an adaptive particle swarm algorithm to optimize the parameters of an LSTM model, wherein the signal fingerprint database includes: signal strength and acquisition time at each location of the test terminal in the indoor scene, detected by multiple access devices set in the indoor scene within a historical time period; The positioning server is further configured to return the target location to the control terminal; Among them, the adaptive particle swarm algorithm is used to optimize the parameters of the LSTM model, including: Initializing a particle swarm of the adaptive particle swarm algorithm to obtain an initial particle swarm, wherein the initial particle swarm includes a plurality of particles, each particle being a parameter combination of the LSTM model; Determine the objective function of the LSTM model as the fitness function of the adaptive particle swarm algorithm, and the objective function is an accuracy evaluation function; Configuring the parameter combination corresponding to each particle in the LSTM model, inputting samples in a sample set into the LSTM model for prediction, and calculating the fitness of each particle using the fitness function based on the predicted value and actual position of the sample to obtain the fitness value of each particle, wherein the sample set is obtained by standardizing the signal fingerprint database; The speed and position of each particle are updated according to the fitness value of each particle, and the fitness of each updated particle is recalculated until the preset number of iterations is reached to obtain the target particle swarm; The parameter combination corresponding to the particle with the largest fitness value in the target particle swarm is configured as the optimal parameter combination into the LSTM model to complete parameter optimization.

Images