CN116761249B - Indoor positioning method, fingerprint library construction method, electronic equipment and storage medium - Google Patents

Abstract

The present application discloses an indoor positioning method, a method for constructing a fingerprint library, an electronic device and a storage medium, and belongs to the field of terminal positioning technology. The method includes: in response to the positioning operation of the positioning device, obtaining the trajectory detection data of the positioning device; based on the trajectory detection data, determining the floor where the positioning device is located through a first classification model; wherein the first classification model is obtained by iteratively training the second classification model based on the first fingerprint library in advance, the first fingerprint library is obtained by expanding the data of the second fingerprint library through the second classification model based on the fingerprint test set, the second classification model is obtained by training the initial classification model based on the second fingerprint library, and the second fingerprint library includes multiple sample trajectory detection data, and each sample trajectory detection data in the multiple sample trajectory detection data has a floor label. The present application expands the fingerprint library while performing model training, thereby improving the reliability of the fingerprint library and the positioning accuracy of the classification model.

Description

Indoor positioning method, fingerprint library construction method, electronic device and storage medium

Technical Field

The present application relates to the field of terminal positioning technology, and in particular to an indoor positioning method, a fingerprint library construction method, an electronic device and a storage medium.

Background Art

With the development of terminal positioning technology, indoor positioning technology is becoming more and more popular. There are many kinds of indoor positioning technologies, such as wireless fidelity (WiFi) fingerprint positioning technology. WiFi fingerprint positioning technology is a technology that allows positioning devices to determine location information based on scanned WiFi signals. In one application scenario, WiFi fingerprint positioning technology can be used to determine the floor where the positioning device is located.

WiFi fingerprint positioning technology includes an offline test phase and an online positioning phase. The key to the offline test phase is to establish a fingerprint library, that is, to establish a correspondence between floors and sample trajectory detection data, where the sample trajectory detection data includes the WiFi signal strength detected at each trajectory point in multiple trajectory points. Afterwards, the fingerprint library can be used for model training to obtain a classification model. In this way, in the online positioning phase, positioning can be performed through the classification model based on the trajectory detection data collected by the positioning device.

However, since the sample trajectory detection data in the fingerprint library is usually collected manually, in order to train the classification model and to test the positioning performance of the trained classification model, the manually collected sample trajectory detection data needs to be used not only to build the fingerprint library, but also to build a fingerprint test set, resulting in a small number of sample trajectory data in the fingerprint library. Therefore, it is difficult to build a reliable fingerprint library, which makes the classification accuracy of the classification model trained based on the fingerprint library low, which in turn affects the accuracy of positioning in the online positioning stage.

Summary of the invention

The present application provides an indoor positioning method, a fingerprint library construction method, an electronic device and a storage medium, which can be used to improve the problem of unreliable fingerprint library in related technologies, resulting in low classification accuracy of classification models. The technical solution is as follows:

In a first aspect, an indoor positioning method is provided, the method comprising:

In response to a positioning operation on a positioning device, acquiring trajectory detection data of the positioning device;

Based on the trajectory detection data, determining the floor where the positioning device is located by using a first classification model;

The first classification model is obtained by iteratively training the second classification model based on the first fingerprint library in advance, the first fingerprint library is obtained by expanding the data of the second fingerprint library through the second classification model based on the fingerprint test set, the second classification model is obtained by training the initial classification model based on the second fingerprint library, and the second fingerprint library includes multiple sample trajectory detection data, and each of the multiple sample trajectory detection data has a floor label.

In this way, when indoor positioning is required, the trajectory detection data of the positioning device can be obtained, and then the floor where the positioning device is located can be determined through the first classification model based on the trajectory detection data. Since the first classification model is trained based on the first fingerprint library after data expansion, and the first fingerprint library is obtained by data expansion of the second fingerprint library based on the fingerprint test set, the data volume in the first fingerprint library is relatively reliable, and thus the positioning accuracy of the first classification model is also improved.

As an example of the present application, the first fingerprint library is obtained by traversing the sample trajectory detection data in the fingerprint test set through the second classification model. During the traversal process, the sample trajectory detection data whose classification result confidence is greater than or equal to the confidence threshold is added to the second fingerprint library.

In this way, during the process of testing the positioning performance of the second classification model based on the fingerprint test set, the data volume of the second fingerprint library can also be expanded, thereby ensuring the data volume of the first fingerprint library and improving the reliability of the first fingerprint library.

As an example of the present application, the second fingerprint library is obtained after data expansion processing is performed on the initial fingerprint library based on the initial fingerprint library and the fingerprint test set, and the initial fingerprint library is obtained by randomly selecting a second number of flat layer sample trajectory detection data from the acquired first number of sample trajectory detection data.

In this way, by randomly selecting the second number of flat-layer sample trajectory detection data, it is ensured that no cross-layer sample trajectory detection data will appear in the initial fingerprint library, thereby ensuring the reliability of the initial fingerprint library.

As an example of the present application, the fingerprint test set is a set consisting of remaining sample trajectory detection data after randomly selecting from the first number of sample trajectory detection data.

In this way, the fingerprint test set and the initial fingerprint library are derived from the same batch of sample trajectory detection data, thereby improving the convenience of constructing the initial fingerprint library and the fingerprint test set.

In a second aspect, a method for constructing a fingerprint library is provided, the method comprising:

Obtain a second fingerprint library and a fingerprint test set, wherein the second fingerprint library includes a plurality of sample trajectory detection data, and each of the plurality of sample trajectory detection data has a floor label;

Iteratively training the initial classification model based on the second fingerprint library to obtain a second classification model;

Performing traversal processing on the sample trajectory detection data in the fingerprint test set by using the second classification model;

During the traversal process, the sample trajectory detection data whose classification result confidence is greater than or equal to the confidence threshold is added to the second fingerprint library;

When the traversal is completed, the second classification model is iteratively trained based on the first fingerprint library to obtain a first classification model. The first fingerprint library is obtained by expanding the second fingerprint library during the traversal process. The first classification model can determine the floor where the positioning device is located based on the trajectory detection data of any positioning device.

In this way, since the first classification model is trained based on the first fingerprint library that has undergone data expansion, and the first fingerprint library is obtained by data expansion of the second fingerprint library based on the fingerprint test set, the amount of data in the first fingerprint library is more reliable, and thus the positioning accuracy of the first classification model is also improved.

As an example of the present application, the classification result includes a floor label and a probability value corresponding to the floor label;

In the traversal process, adding the sample trajectory detection data whose classification result confidence is greater than or equal to the confidence threshold to the second fingerprint library includes:

During the traversal process, according to the probability value in the classification result corresponding to the first sample trajectory detection data, the confidence of the classification result corresponding to the first sample trajectory detection data is determined, and the first sample trajectory detection data is any sample trajectory detection data in the fingerprint test set currently traversed;

When the confidence of the classification result corresponding to the first sample trajectory detection data is greater than or equal to the confidence threshold, updating the floor label of the first sample trajectory detection data to the floor label in the classification result corresponding to the first sample trajectory detection data;

The first sample trajectory detection data after label update is added to the second fingerprint library.

In this way, in the process of testing the second classification model based on the fingerprint test set, the data of the second fingerprint library can also be expanded, so that the model training and the fingerprint library construction can complement each other.

As an example of the present application, the obtaining of the second fingerprint library and the fingerprint test set includes:

Acquire a first number of sample trajectory detection data, where the first number of sample trajectory detection data at least includes level layer sample trajectory detection data;

Randomly selecting a second number of flat layer sample trajectory detection data from the first number of sample trajectory detection data to obtain an initial fingerprint library;

Determine a set consisting of remaining sample trajectory detection data after randomly selecting the first number of sample trajectory detection data as the fingerprint test set;

Based on the initial fingerprint library and the fingerprint test set, data expansion processing is performed on the initial fingerprint library to obtain the second fingerprint library.

In this way, the initial fingerprint library is expanded through the fingerprint test set, thereby ensuring the data volume of the sample trajectory detection data in the second fingerprint library and improving the reliability of the second fingerprint library.

As an example of the present application, the data expansion processing of the initial fingerprint library based on the initial fingerprint library and the fingerprint test set includes:

Traversing each sample trajectory detection data in the fingerprint test set in turn;

Determine the Euclidean distance and the Jaccard distance between the second sample trajectory detection data and each sample trajectory detection data in the initial fingerprint library, wherein the second sample trajectory detection data is the currently traversed sample trajectory detection data;

In the case where there is at least one third sample trajectory detection data in the initial fingerprint library, determining the third sample trajectory detection data having the greatest similarity to the second sample trajectory detection data from the at least one third sample trajectory detection data, wherein the third sample trajectory detection data refers to the sample trajectory detection data having a Euclidean distance with the second sample trajectory detection data less than a first distance threshold and a Jaccard distance with the second sample trajectory detection data less than a second distance threshold;

Updating the floor label of the second sample trajectory detection data to the determined floor label of the third sample trajectory detection data;

The second sample trajectory detection data after the label is updated is added to the initial fingerprint library.

In this way, by determining the third sample trajectory detection data having the greatest similarity to the second sample trajectory detection data from at least one third sample trajectory detection data, the rationality of expanding the initial fingerprint library is ensured.

As an example of the present application, when there is at least one third sample trajectory detection data in the initial fingerprint library, determining the third sample trajectory detection data having the greatest similarity to the second sample trajectory detection data from the at least one third sample trajectory detection data includes:

In the case where there is at least one third sample trajectory detection data in the initial fingerprint library, obtaining the third sample trajectory detection data having the smallest Euclidean distance with the second sample trajectory data from the at least one third sample trajectory detection data; or,

In the case that at least one third sample trajectory detection data exists in the initial fingerprint library, the third sample trajectory detection data having the smallest Jaccard distance with the second sample trajectory data is obtained from the at least one third sample trajectory detection data.

In this way, by using the Euclidean distance or the Jaccard distance, the third sample trajectory detection data having the greatest similarity to the second sample trajectory detection data is selected from at least one third sample trajectory detection data, so that the selected third sample trajectory detection data can be more accurate.

As an example of the present application, the data expansion processing of the initial fingerprint library based on the initial fingerprint library and the fingerprint test set includes:

Traversing each sample trajectory detection data in the fingerprint test set in turn;

Determine the similarity between the second sample trajectory detection data and each sample trajectory detection data in the initial fingerprint library to obtain multiple similarities, wherein the second sample trajectory detection data is the currently traversed sample trajectory detection data;

When the maximum similarity among the multiple similarities is greater than or equal to the similarity threshold, updating the floor label of the second sample trajectory detection data to the floor label of the sample trajectory detection data corresponding to the maximum similarity;

The second sample trajectory detection data after the label is updated is added to the initial fingerprint library.

In this way, the data expansion processing is performed on the initial fingerprint library in different ways, thereby increasing the richness of the determined data expansion methods.

As an example of the present application, before randomly selecting a second number of flat layer sample trajectory detection data from the first number of sample trajectory detection data to obtain an initial fingerprint library, the method further includes:

In a case where the first number of sample trajectory detection data includes sample trajectory detection data whose number of trajectory points is greater than a number threshold, segmenting the sample trajectory detection data whose number of trajectory points is greater than the number threshold to obtain a third number of sample trajectory detection data, wherein the number of trajectory points of each sample trajectory detection data in the third number of sample trajectory detection data is less than or equal to the number threshold;

The randomly selecting a second number of flat layer sample trajectory detection data from the first number of sample trajectory detection data to obtain an initial fingerprint library includes:

The second number of flat layer sample trajectory detection data is randomly selected from the third number of sample trajectory detection data to obtain the initial fingerprint library.

In this way, by segmenting the longer sample trajectory detection data, the number of trajectory points included in the longer sample trajectory detection data is reduced, thereby reducing the amount of subsequent calculations based on the sample trajectory detection data.

As an example of the present application, when the first number of sample trajectory detection data includes sample trajectory detection data whose number of trajectory points is greater than the number threshold, segmenting the sample trajectory detection data whose number of trajectory points is greater than the number threshold includes:

In a case where the first number of sample trajectory detection data includes sample trajectory detection data whose number of trajectory points is greater than the number threshold, segmenting the sample trajectory detection data whose number of trajectory points is greater than the number threshold at intervals of a segmentation value to obtain a fourth number of sample trajectory detection data, wherein the segmentation value is a value less than or equal to the number threshold;

In the case that the fourth amount of sample trajectory detection data includes sample trajectory detection data whose number of trajectory points is smaller than the division value, the sample trajectory detection data whose number of trajectory points is smaller than the division value is deleted.

In this way, by dividing the sample trajectory detection data whose number of trajectory points is greater than the quantity threshold into intervals, not only can the number of sample trajectory detection data be expanded, but also the amount of calculation in subsequent model training can be reduced and the accuracy of model training can be improved.

As an example of the present application, after traversing the sample trajectory detection data in the fingerprint test set by the second classification model, the method further includes:

Obtaining the correct amount of sample trajectory detection data in the fingerprint test set classified by the second classification model, and the amount of sample trajectory detection data added to the second fingerprint library after traversing the sample trajectory detection data in the fingerprint test set through the second classification model;

The classification accuracy is obtained by dividing the correct amount by the total number of samples in the fingerprint test set, where the total number of samples in the fingerprint test set is the number of sample trajectory detection data included in the fingerprint test set before the sample trajectory detection data in the fingerprint test set is added to the second fingerprint library through the second classification model;

The added amount is divided by the total number of samples to obtain the track utilization rate. The classification accuracy and the track utilization rate are used to evaluate the positioning performance of the first classification model.

In this way, by determining the classification accuracy of the second classification model and the trajectory utilization rate, the positioning performance of the second classification model can be evaluated using specific data, thereby improving the accuracy of evaluating the positioning performance of the second classification model.

In a third aspect, an indoor positioning device is provided, wherein the indoor positioning device has the function of implementing the indoor positioning method in the first aspect. The indoor positioning device includes at least one module, and the at least one module is used to implement the indoor positioning method provided in the first aspect. The indoor positioning device may include:

An acquisition module, configured to acquire trajectory detection data of the positioning device in response to a positioning operation on the positioning device;

A determination module, configured to determine the floor where the positioning device is located by using a first classification model based on the trajectory detection data;

The first classification model is obtained by iteratively training the second classification model based on the first fingerprint library in advance, the first fingerprint library is obtained by expanding the data of the second fingerprint library through the second classification model based on the fingerprint test set, the second classification model is obtained by training the initial classification model based on the second fingerprint library, and the second fingerprint library includes multiple sample trajectory detection data, and each of the multiple sample trajectory detection data has a floor label.

As an example of the present application, the first fingerprint library is obtained by traversing the sample trajectory detection data in the fingerprint test set through the second classification model. During the traversal process, the sample trajectory detection data whose classification result confidence is greater than or equal to the confidence threshold is added to the second fingerprint library.

As an example of the present application, the second fingerprint library is obtained after data expansion processing is performed on the initial fingerprint library based on the initial fingerprint library and the fingerprint test set, and the initial fingerprint library is obtained by randomly selecting a second number of flat layer sample trajectory detection data from the acquired first number of sample trajectory detection data.

As an example of the present application, the fingerprint test set is a set consisting of remaining sample trajectory detection data after randomly selecting from the first number of sample trajectory detection data.

In a fourth aspect, a fingerprint library construction device is provided, wherein the fingerprint library construction device has the function of implementing the fingerprint library construction method in the second aspect. The fingerprint library construction device includes at least one module, and the at least one module is used to implement the fingerprint library construction method provided in the second aspect. The fingerprint library construction device may include:

A first acquisition module is used to acquire a second fingerprint library and a fingerprint test set, wherein the second fingerprint library includes a plurality of sample trajectory detection data, and each of the plurality of sample trajectory detection data has a floor label;

A first training module, configured to iteratively train the initial classification model based on the second fingerprint library to obtain a second classification model;

A traversal module, used for traversing the sample trajectory detection data in the fingerprint test set through the second classification model;

An adding module, used for adding sample trajectory detection data whose classification result confidence is greater than or equal to the confidence threshold to the second fingerprint library during the traversal process;

The second training module is used to iteratively train the second classification model based on the first fingerprint library to obtain the first classification model when the traversal is completed. The first fingerprint library is obtained by expanding the second fingerprint library during the traversal process. The first classification model can determine the floor where the positioning device is located based on the trajectory detection data of any positioning device.

As an example of the present application, the classification result includes a floor label and a probability value corresponding to the floor label;

The adding module is used for:

During the traversal process, according to the probability value in the classification result corresponding to the first sample trajectory detection data, the confidence of the classification result corresponding to the first sample trajectory detection data is determined, and the first sample trajectory detection data is any sample trajectory detection data in the fingerprint test set currently traversed;

When the confidence of the classification result corresponding to the first sample trajectory detection data is greater than or equal to the confidence threshold, updating the floor label of the first sample trajectory detection data to the floor label in the classification result corresponding to the first sample trajectory detection data;

The first sample trajectory detection data after label update is added to the second fingerprint library.

As an example of the present application, the first acquisition module is used for:

Acquire a first number of sample trajectory detection data, where the first number of sample trajectory detection data at least includes level layer sample trajectory detection data;

Randomly selecting a second number of flat layer sample trajectory detection data from the first number of sample trajectory detection data to obtain an initial fingerprint library;

Determine a set consisting of remaining sample trajectory detection data after randomly selecting the first number of sample trajectory detection data as the fingerprint test set;

Based on the initial fingerprint library and the fingerprint test set, data expansion processing is performed on the initial fingerprint library to obtain the second fingerprint library.

As an example of the present application, the first acquisition module is used for:

Traversing each sample trajectory detection data in the fingerprint test set in turn;

Determine the Euclidean distance and the Jaccard distance between the second sample trajectory detection data and each sample trajectory detection data in the initial fingerprint library, wherein the second sample trajectory detection data is the currently traversed sample trajectory detection data;

In the case where there is at least one third sample trajectory detection data in the initial fingerprint library, determining the third sample trajectory detection data having the greatest similarity to the second sample trajectory detection data from the at least one third sample trajectory detection data, wherein the third sample trajectory detection data refers to the sample trajectory detection data having a Euclidean distance with the second sample trajectory detection data less than a first distance threshold and a Jaccard distance with the second sample trajectory detection data less than a second distance threshold;

Updating the floor label of the second sample trajectory detection data to the determined floor label of the third sample trajectory detection data;

The second sample trajectory detection data after the label is updated is added to the initial fingerprint library.

As an example of the present application, the first acquisition module is used for:

In the case where there is at least one third sample trajectory detection data in the initial fingerprint library, obtaining the third sample trajectory detection data having the smallest Euclidean distance with the second sample trajectory data from the at least one third sample trajectory detection data; or,

In the case that at least one third sample trajectory detection data exists in the initial fingerprint library, the third sample trajectory detection data having the smallest Jaccard distance with the second sample trajectory data is obtained from the at least one third sample trajectory detection data.

As an example of the present application, the first acquisition module is used for:

Traversing each sample trajectory detection data in the fingerprint test set in turn;

Determine the similarity between the second sample trajectory detection data and each sample trajectory detection data in the initial fingerprint library to obtain multiple similarities, wherein the second sample trajectory detection data is the currently traversed sample trajectory detection data;

When the maximum similarity among the multiple similarities is greater than or equal to the similarity threshold, updating the floor label of the second sample trajectory detection data to the floor label of the sample trajectory detection data corresponding to the maximum similarity;

The second sample trajectory detection data after the label is updated is added to the initial fingerprint library.

As an example of the present application, the first acquisition module is further used for:

In a case where the first number of sample trajectory detection data includes sample trajectory detection data whose number of trajectory points is greater than a number threshold, segmenting the sample trajectory detection data whose number of trajectory points is greater than the number threshold to obtain a third number of sample trajectory detection data, wherein the number of trajectory points of each sample trajectory detection data in the third number of sample trajectory detection data is less than or equal to the number threshold;

The second number of flat layer sample trajectory detection data is randomly selected from the third number of sample trajectory detection data to obtain the initial fingerprint library.

As an example of the present application, the first acquisition module is used for:

In a case where the first number of sample trajectory detection data includes sample trajectory detection data whose number of trajectory points is greater than the number threshold, segmenting the sample trajectory detection data whose number of trajectory points is greater than the number threshold at intervals of a segmentation value to obtain a fourth number of sample trajectory detection data, wherein the segmentation value is a value less than or equal to the number threshold;

In the case that the fourth amount of sample trajectory detection data includes sample trajectory detection data whose number of trajectory points is smaller than the division value, the sample trajectory detection data whose number of trajectory points is smaller than the division value is deleted.

As an example of the present application, the device further includes:

A second acquisition module is used to obtain the correct amount of sample trajectory detection data in the fingerprint test set classified by the second classification model, and the amount of sample trajectory detection data added to the second fingerprint library after traversing the sample trajectory detection data in the fingerprint test set through the second classification model;

a first calculation module, configured to obtain the classification accuracy by dividing the correct amount by the total number of samples in the fingerprint test set, where the total number of samples in the fingerprint test set is the number of sample trajectory detection data included in the fingerprint test set before the sample trajectory detection data in the fingerprint test set is added to the second fingerprint library through the second classification model;

The second calculation module is used to divide the added amount by the total sample amount to obtain the trajectory utilization rate, and the classification accuracy and the trajectory utilization rate are used to evaluate the positioning performance of the first classification model.

In a fifth aspect, an electronic device is provided, wherein the structure of the electronic device includes a processor and a memory, wherein the memory is used to store a program that supports the electronic device to execute the indoor positioning method provided in the first aspect, and to store data involved in implementing the indoor positioning method described in the first aspect. Alternatively, the memory is used to store a program that supports the electronic device to execute the method for constructing a fingerprint library provided in the second aspect, and to store data involved in implementing the method for constructing a fingerprint library described in the second aspect; the processor is configured to execute the program stored in the memory. The electronic device may also include a communication bus, which is used to establish a connection between the processor and the memory.

In a sixth aspect, a computer-readable storage medium is provided, wherein instructions are stored in the computer-readable storage medium, and when the computer-readable storage medium is run on a computer, the computer executes the indoor positioning method described in the first aspect, or the fingerprint library construction method described in the second aspect.

In a seventh aspect, a computer program product comprising instructions is provided, which, when executed on a computer, enables the computer to execute the indoor positioning method described in the first aspect, or enables the computer to execute the fingerprint library construction method described in the second aspect.

The technical effects obtained by the above-mentioned second, third, fourth, fifth, sixth and seventh aspects are similar to the technical effects obtained by the corresponding technical means in the above-mentioned first aspect, and will not be repeated here.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a schematic diagram of an application scenario for collecting sample trajectory detection data provided by an embodiment of the present application;

FIG2 is a schematic diagram of an application scenario of indoor positioning provided by an embodiment of the present application;

FIG3 is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application;

FIG4 is a block diagram of a software system of an electronic device provided in an embodiment of the present application;

FIG5 is a schematic diagram of an application scenario of indoor positioning provided by an embodiment of the present application;

FIG6 is a schematic diagram of another application scenario of indoor positioning provided by an embodiment of the present application;

FIG7 is a schematic diagram of another application scenario of indoor positioning provided by an embodiment of the present application;

FIG8 is a schematic diagram of another application scenario of indoor positioning provided by an embodiment of the present application;

FIG9 is a schematic flow chart of an indoor positioning method provided by an embodiment of the present application;

FIG10 is a schematic diagram of a method for constructing a fingerprint library according to an embodiment of the present application;

FIG11 is a schematic diagram of different types of sample trajectory detection data provided by an embodiment of the present application;

FIG12 is a flow chart of a method for performing data expansion processing on an initial fingerprint library provided in an embodiment of the present application;

FIG13 is a schematic flow chart of another method for constructing a fingerprint library provided in an embodiment of the present application;

FIG14 is a schematic diagram of the structure of an indoor positioning device provided in an embodiment of the present application;

FIG. 15 is a schematic diagram of the structure of a fingerprint library construction device provided in an embodiment of the present application.

DETAILED DESCRIPTION

In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below in conjunction with the accompanying drawings.

It should be understood that the "multiple" mentioned in this application refers to two or more. In the description of this application, unless otherwise specified, "/" means or, for example, A/B can mean A or B; "and/or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships, for example, A and/or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, in order to facilitate the clear description of the technical solution of this application, the words "first" and "second" are used to distinguish between the same or similar items with basically the same functions and effects. Those skilled in the art can understand that the words "first" and "second" do not limit the quantity and execution order, and the words "first" and "second" do not limit them to be different.

References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

In an application scenario, when an electronic device needs to locate itself, if the electronic device is in an indoor environment, the electronic device can be located by indoor positioning technology, for example, the electronic device can be located by WiFi fingerprint positioning technology. When the electronic device is in an indoor environment, multiple sample trajectory detection data collected manually and the floor label corresponding to each sample trajectory detection data can be obtained in advance. Exemplarily, referring to FIG1, at least one collector uses a data acquisition device to collect sample trajectory detection data on different floors, and sends the collected sample trajectory detection data and the corresponding floor to the electronic device. For ease of explanation, the accompanying drawings of the embodiment of the present application are illustrated by taking 4 floors and one collector on each floor as an example. After obtaining multiple sample trajectory detection data and the floor label corresponding to each sample trajectory detection data, the electronic device can select a fixed number of sample trajectory detection data from the sample trajectory data corresponding to each floor, thereby forming a fingerprint library, and use the fingerprint library for model training to obtain a classification model. Generally, the electronic device can also obtain a fingerprint test set, based on which the positioning performance of the classification model can be tested, that is, whether the classification model can be used for indoor positioning.

However, since the manually collected sample trajectory detection data is not only needed to build a fingerprint library, but also needs to build a fingerprint test set, the number of sample trajectory data in the fingerprint library is small, so it is difficult to build a reliable fingerprint library, which makes the classification accuracy of the classification model trained based on the fingerprint library low, thereby affecting the accuracy of positioning in the online positioning stage. For example, referring to Figure 2, if the electronic device is on the 3rd floor, the electronic device can obtain trajectory detection data of any time period on the 3rd floor, and locate it through the classification model based on the trajectory detection data. The positioning result of the classification model should originally be the 3rd floor, but because the classification model is trained based on a fingerprint library with a small number of sample trajectory data, the fingerprint library is unreliable, resulting in inaccurate positioning of the classification model, causing the classification model to output a positioning result of the 4th floor, and the electronic device displays the positioning result "Currently on the 4th floor" in the positioning interface.

In order to improve the positioning accuracy of the classification model, an embodiment of the present application provides an indoor positioning method, in which the electronic device can obtain the trajectory detection data of the positioning device when indoor positioning is required, and then determine the floor where the positioning device is located through the first classification model based on the trajectory detection data. Since the first classification model is trained based on the first fingerprint library after data expansion, and the first fingerprint library is obtained by data expansion of the second fingerprint library based on the fingerprint test set, the amount of data in the first fingerprint library is more reliable, and thus the positioning accuracy of the first classification model is also improved. That is, the electronic device in the embodiment of the present application expands the amount of data in the fingerprint library, improves the reliability of the fingerprint library, and thus improves the positioning accuracy of the classification model.

Before explaining the indoor positioning method provided in the embodiment of the present application in detail, the electronic device involved in the embodiment of the present application is first described. The method provided in the embodiment of the present application can be performed by an electronic device, and the electronic device can be configured with a wifi signal detection function and a positioning function. Further, the electronic device can be installed with applications such as navigation, social networking, shopping, etc., and these applications can trigger the electronic device to locate during use. As an example and not a limitation, the electronic device can be but not limited to a mobile phone, a wearable smart device, a digital camera, a tablet computer, a desktop, a laptop, a handheld computer, a notebook computer, a vehicle-mounted device, an ultra-mobile personal computer (ultra-mobilepersonal computer, UMPC), a netbook, a cellular phone, a personal digital assistant (personal digitalassistant, PDA), an augmented reality (augmented reality, AR)\virtual reality (virtual reality, VR) device, a mobile phone, a smart appliance, a server, etc., and the embodiment of the present application is not limited to this.

Fig. 3 is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Referring to Fig. 3, the electronic device 100 may include a processor 110, an external memory interface 120, an internal memory 121, a universal serial bus (USB) interface 130, a charging management module 140, a power management module 141, a battery 142, an antenna 1, an antenna 2, a mobile communication module 150, a wireless communication module 160, an audio module 170, a speaker 170A, a receiver 170B, a microphone 170C, an earphone interface 170D, a sensor module 180, a button 190, a motor 191, an indicator 192, a camera 193, a display screen 194, and a subscriber identification module (SIM) card interface 195, etc. Among them, the sensor module 180 may include a pressure sensor 180A, a gyroscope sensor 180B, an air pressure sensor 180C, a magnetic sensor 180D, an acceleration sensor 180E, a distance sensor 180F, a proximity light sensor 180G, a fingerprint sensor 180H, a temperature sensor 180J, a touch sensor 180K, an ambient light sensor 180L, a bone conduction sensor 180M, etc.

It is to be understood that the structure illustrated in the embodiment of the present application does not constitute a specific limitation on the electronic device 100. In other embodiments of the present application, the electronic device 100 may include more or fewer components than shown in the figure, or combine some components, or split some components, or arrange the components differently. The components shown in the figure may be implemented in hardware, software, or a combination of software and hardware.

The processor 110 may include one or more processing units, for example, the processor 110 may include an application processor (AP), a modem processor, a graphics processor (GPU), an image signal processor (ISP), a controller, a memory, a video codec, a digital signal processor (DSP), a baseband processor, and/or a neural-network processing unit (NPU), etc. Different processing units may be independent devices or integrated into one or more processors.

The controller may be the nerve center and command center of the electronic device 100. The controller may generate an operation control signal according to the instruction operation code and the timing signal to complete the control of fetching and executing instructions.

The processor 110 may also be provided with a memory for storing instructions and data. In some embodiments, the memory in the processor 110 is a cache memory. The memory may store instructions or data that the processor 110 has just used or cyclically used. If the processor 110 needs to use the instruction or data again, it may be directly called from the memory. This avoids repeated access, reduces the waiting time of the processor 110, and thus improves the efficiency of the system.

In some embodiments, the processor 110 may include one or more interfaces, such as an inter-integrated circuit (I2C) interface, an inter-integrated circuit sound (I2S) interface, a pulse code modulation (PCM) interface, a universal asynchronous receiver/transmitter (UART) interface, a mobile industry processor interface (MIPI), a general-purpose input/output (GPIO) interface, a subscriber identity module (SIM) interface, and/or a universal serial bus (USB) interface, etc.

It is understandable that the interface connection relationship between the modules illustrated in the embodiment of the present application is only a schematic illustration and does not constitute a structural limitation on the electronic device 100. In other embodiments of the present application, the electronic device 100 may also adopt different interface connection methods in the above embodiments, or a combination of multiple interface connection methods.

The wireless communication function of the electronic device 100 can be implemented through the antenna 1, the antenna 2, the mobile communication module 150, the wireless communication module 160, the modem processor and the baseband processor.

Antenna 1 and antenna 2 are used to transmit and receive electromagnetic wave signals. Each antenna in the electronic device 100 can be used to cover a single or multiple communication frequency bands. Different antennas can also be reused to improve the utilization of the antennas. For example, antenna 1 can be reused as a diversity antenna for a wireless local area network. In some other embodiments, the antenna can be used in combination with a tuning switch.

The mobile communication module 150 can provide solutions for wireless communications including 2G/3G/4G/5G, etc., applied to the electronic device 100. The mobile communication module 150 may include at least one filter, a switch, a power amplifier, a low noise amplifier (LNA), etc. The mobile communication module 150 can receive electromagnetic waves from the antenna 1, and filter, amplify, and process the received electromagnetic waves, and transmit them to the modulation and demodulation processor for demodulation. The mobile communication module 150 can also amplify the signal modulated by the modulation and demodulation processor, and convert it into electromagnetic waves for radiation through the antenna 1. In some embodiments, at least some of the functional modules of the mobile communication module 150 can be set in the processor 110. In some embodiments, at least some of the functional modules of the mobile communication module 150 can be set in the same device as at least some of the modules of the processor 110.

The modem processor may include a modulator and a demodulator. Among them, the modulator is used to modulate the low-frequency baseband signal to be sent into a medium-high frequency signal. The demodulator is used to demodulate the received electromagnetic wave signal into a low-frequency baseband signal. The demodulator then transmits the demodulated low-frequency baseband signal to the baseband processor for processing. After the low-frequency baseband signal is processed by the baseband processor, it is passed to the application processor. The application processor outputs a sound signal through an audio device (not limited to a speaker 170A, a receiver 170B, etc.), or displays an image or video through a display screen 194. In some embodiments, the modem processor may be an independent device. In other embodiments, the modem processor may be independent of the processor 110 and be set in the same device as the mobile communication module 150 or other functional modules.

The wireless communication module 160 can provide wireless communication solutions including wireless local area networks (WLAN) (such as wireless fidelity (Wi-Fi) networks), bluetooth (BT), global navigation satellite system (GNSS), frequency modulation (FM), near field communication (NFC), infrared (IR), etc., which are applied to the electronic device 100. The wireless communication module 160 can be one or more devices integrating at least one communication processing module. The wireless communication module 160 receives electromagnetic waves via the antenna 2, modulates the frequency of the electromagnetic wave signal and performs filtering, and sends the processed signal to the processor 110. The wireless communication module 160 can also receive the signal to be sent from the processor 110, modulate the frequency of it, amplify it, and convert it into electromagnetic waves for radiation through the antenna 2.

In some embodiments, the antenna 1 of the electronic device 100 is coupled to the mobile communication module 150, and the antenna 2 is coupled to the wireless communication module 160, so that the electronic device 100 can communicate with the network and other devices through wireless communication technology. The wireless communication technology may include global system for mobile communications (GSM), general packet radio service (GPRS), code division multiple access (CDMA), wideband code division multiple access (WCDMA), time-division code division multiple access (TD-SCDMA), long term evolution (LTE), BT, GNSS, WLAN, NFC, FM, and/or IR technology, etc. GNSS may include the global positioning system (GPS), the global navigation satellite system (GLONASS), the Beidou navigation satellite system (BDS), the quasi-zenith satellite system (QZSS) and/or the satellite based augmentation system (SBAS).

The electronic device 100 implements the display function through a GPU, a display screen 194, and an application processor. The GPU is a microprocessor for image processing, which connects the display screen 194 and the application processor. The GPU is used to perform mathematical and geometric calculations for graphics rendering. The processor 110 may include one or more GPUs that execute program instructions to generate or change display information.

The digital signal processor is used to process digital signals, and can process not only digital image signals but also other digital signals. For example, when the electronic device 100 is selecting a frequency point, the digital signal processor is used to perform Fourier transform on the frequency point energy.

The external memory interface 120 can be used to connect an external memory card, such as a Micro SD card, to expand the storage capacity of the electronic device 100. The external memory card communicates with the processor 110 through the external memory interface 120 to implement a data storage function, such as storing music, video and other files in the external memory card.

The internal memory 121 can be used to store computer executable program codes, which include instructions. The processor 110 executes various functional applications and data processing of the electronic device 100 by running the instructions stored in the internal memory 121. The internal memory 121 may include a program storage area and a data storage area. Among them, the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc. The data storage area may store data created by the electronic device 100 during use (such as audio data, a phone book, etc.), etc. In addition, the internal memory 121 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, a universal flash storage (UFS), etc.

The air pressure sensor 180C is used to measure air pressure. In some embodiments, the electronic device 100 calculates the altitude through the air pressure value measured by the air pressure sensor 180C to assist positioning and navigation.

Next, the software system of the electronic device 100 will be described.

The software system of the electronic device 100 may adopt a layered architecture, an event-driven architecture, a micro-core architecture, a micro-service architecture, or a cloud architecture. The embodiment of the present application takes the Android system of the layered architecture as an example to exemplify the software system of the electronic device 100.

FIG4 is a block diagram of a software system of an electronic device 100 provided in an embodiment of the present application. Referring to FIG4 , the layered architecture divides the software into several layers, each layer having a clear role and division of labor. The layers communicate with each other through software interfaces. In some embodiments, the Android system is divided into four layers, namely, from top to bottom, the application layer, the application framework layer, the Android runtime (Android runtime) and the system layer, and the kernel layer.

The application layer may include a series of application packages. As shown in FIG4 , the application package may include applications such as map, navigation, camera, gallery, calendar, call, WLAN, Bluetooth, music, video, etc.

As an example, the application layer may also include a WiFi management module, which is used to manage WiFi-related functions, such as turning on Wifi, disconnecting Wifi, viewing the Wifi list, dynamically refreshing the Wifi list, dynamically pulling down to refresh the Wifi list, connecting to a specified network, disconnecting the network, etc.

The application framework layer provides an application programming interface (API) and a programming framework for the application programs of the application layer. The application framework layer includes some predefined functions. As shown in FIG4 , the application framework layer may include a window manager, a content provider, a view system, a phone manager, a resource manager, a notification manager, etc. The window manager is used to manage window programs. The window manager can obtain the size of the display screen, determine whether there is a status bar, lock the screen, capture the screen, etc. The content provider is used to store and obtain data and make these data accessible to applications. These data may include video, images, audio, dialed and received calls, browsing history and bookmarks, phone books, track detection data, etc. The view system includes visual controls, such as controls for displaying text, controls for displaying pictures, etc. The view system can be used to construct the display interface of the application, and the display interface can be composed of one or more views, for example, including a view for displaying a text message notification icon, a view for displaying text, and a view for displaying pictures. The phone manager is used to provide communication functions of the electronic device 100, such as management of call status (including connecting, hanging up, etc.). The resource manager provides various resources for applications, such as localized strings, icons, images, layout files, video files, etc. The notification manager enables applications to display notification information in the status bar. It can be used to convey notification-type messages and can disappear automatically after a short stay without user interaction. For example, the notification manager is used to notify the completion of downloads, message reminders, etc. The notification manager can also be a notification that appears in the system top status bar in the form of a chart or scroll bar text, such as notifications of applications running in the background. The notification manager can also be a notification that appears on the screen in the form of a dialog window, such as a text message prompt in the status bar, a reminder sound, an electronic device vibrating, an indicator light flashing, etc.

As an example, the application framework layer may also include an algorithm library and a data receiving module.

As an example, the algorithm module is used to store various algorithm models, which include a first classification model for realizing indoor positioning; when indoor positioning is required, the algorithm module can call the stored first classification module, and when the trajectory detection data of the positioning device is obtained, the first classification module is used to determine the floor where the positioning device is located based on the trajectory detection data. Exemplarily, if the algorithm module receives a positioning request triggered by an application such as a map or navigation, the algorithm module can call the first classification module, and when the trajectory detection data of the positioning device is obtained from the content provider, the first classification module is used to determine the floor where the positioning device is located based on the trajectory detection data through the set positioning algorithm.

As an example, the algorithm module can also be used for model training, and when performing model training, the algorithm module can receive a certain number of sample trajectory detection data sent by the data acquisition module, and then perform model training based on the acquired sample trajectory detection data.

As an example, the data acquisition module may be used to receive sample trajectory detection data sent by other electronic devices, and send the received sample trajectory detection data to the algorithm module.

As an example, the data acquisition module can also receive WiFi signals periodically collected by the WiFi collector in the hardware layer, and determine the WiFi signal strength of each received WiFi signal as the WiFi signal strength corresponding to a trajectory point to obtain trajectory detection data, and send the trajectory detection data to the algorithm module.

As an example, the data acquisition module may also be used to store the acquired trajectory detection data in the content provider.

Android Runtime includes core libraries and virtual machines. Android runtime is responsible for scheduling and management of the Android system. The core library consists of two parts: one is the function that the Java language needs to call, and the other is the Android core library. The application layer and the application framework layer run in the virtual machine. The virtual machine executes the Java files of the application layer and the application framework layer as binary files. The virtual machine is used to perform object life cycle management, stack management, thread management, security and exception management, and garbage collection.

The system library can include multiple functional modules, such as: surface manager, media library, 3D graphics processing library (such as: OpenGL ES), 2D graphics engine (such as: SGL), etc. The surface manager is used to manage the display subsystem and provide fusion of 2D and 3D layers for multiple applications. The media library supports playback and recording of a variety of commonly used audio and video formats, as well as static image files, etc. The media library can support a variety of audio and video encoding formats, such as: MPEG4, H.264, MP3, AAC, AMR, JPG, PNG, etc. The 3D graphics processing library is used to implement 3D graphics drawing, image rendering, synthesis, and layer processing, etc. The 2D graphics engine is a drawing engine for 2D drawing.

The kernel layer is the layer between hardware and software. The kernel layer includes at least display driver, camera driver, audio driver, sensor driver, and WiFi driver;

As an example, the WiFi driver is used to drive a hardware device WiFi collector to periodically collect WiFi signals.

The following is an illustrative description of the workflow of the software and hardware of the electronic device 100 in conjunction with capturing a photo scene.

When the touch sensor 180K receives a touch operation, the corresponding hardware interrupt is sent to the kernel layer. The kernel layer processes the touch operation into an original input event (including touch coordinates, timestamp of the touch operation, and other information). The original input event is stored in the kernel layer. The application framework layer obtains the original input event from the kernel layer and identifies the control corresponding to the original input event. For example, if the touch operation is a single-click operation and the control corresponding to the single-click operation is the control of the camera application icon, the camera application calls the interface of the application framework layer, starts the camera application, and then calls the kernel layer to start the camera driver to capture static images or videos through the camera 193.

For ease of understanding, before introducing the method provided in the embodiment of the present application in detail, based on the execution entity provided in the above embodiment, the application scenarios involved in the embodiment of the present application are introduced below.

Please refer to Figure 5, which is a schematic diagram of an application scenario according to an exemplary embodiment. In a possible scenario, when a user visits a factory building, in order to prevent the user from entering a floor that is not allowed to enter, the user can carry a mobile phone dedicated to positioning. When the user carries the mobile phone dedicated to positioning, the mobile phone can perform positioning operations on itself in real time or at specified intervals, and warn the user when it is located that the user has entered a floor that is not allowed to enter. Exemplarily, in response to the positioning operation on itself, the mobile phone can obtain the trajectory detection data of the user within a specified time period, and based on the trajectory detection data, determine through the first classification model that the current floor of the mobile phone is the 3rd floor. If the 3rd floor is a floor that the user is not allowed to enter, the mobile phone can vibrate, ring, text prompt and/or voice prompt to prompt the user to leave the 3rd floor. For example, the mobile phone can display "Please leave the 3rd floor as soon as possible" on the positioning interface.

It should be noted that the specified time interval can be set in advance, for example, the specified time interval can be 3 minutes, 5 minutes or 10 minutes, etc. The specified time period can be set in advance according to the needs, and the specified time period is different depending on the positioning object. For example, in the case where the electronic device needs to perform self-positioning, the specified time period is a time period for collecting trajectory detection data before receiving the positioning operation moment and the closest to the receiving positioning operation moment, and the time period can be 5 minutes, 3 minutes or 1 minute, etc. In the case where the electronic device needs to locate other devices, the specified time period is a time period for collecting trajectory detection data before the receiving moment of the trajectory acquisition request sent by the electronic device and the closest to the receiving moment, and the time period can be 5 minutes, 3 minutes or 1 minute, etc.

As an example, since the electronic device can scan the WiFi signal of the current environment according to its own signal scanning cycle, and obtain a trajectory point and the signal strength of the WiFi signal detected at the trajectory point in each scan, the electronic device can determine the signal strength of the WiFi signal detected at multiple trajectory points in any time period and each of the multiple trajectory points as trajectory detection data.

As an example of the present application, a plurality of intelligent cleaning devices are usually placed in a shopping mall building, and the plurality of intelligent cleaning devices are used to clean the environment and sanitation of the shopping mall building. In order to facilitate management and planning, in order to clearly know the number of intelligent cleaning devices on each floor, the manager can use electronic devices such as mobile phones, laptops, desktop computers, etc. to perform indoor positioning of each intelligent cleaning device. In FIG. 6 of the embodiment of the present application, the electronic device is taken as an example of a mobile phone for illustration. Referring to FIG. 6 (a), the manager can trigger the positioning operation of the intelligent cleaning device in the management interface displayed by the mobile phone, for example, the manager can click the "distribution of cleaning equipment" option in the management interface; in response to the manager's click operation on the "distribution of cleaning equipment", the mobile phone can send a trajectory acquisition request to each intelligent cleaning device; each intelligent cleaning device, when receiving the trajectory acquisition request, can send the trajectory detection data collected within a specified time period to the electronic device; the electronic device receives the trajectory detection data sent by each intelligent cleaning device, and based on the trajectory detection data of each intelligent cleaning device, determines the floor where each intelligent cleaning device is located through the first classification model; afterwards, referring to FIG. 6 (b), the electronic device can count the number of intelligent cleaning devices placed on each floor, and display the number of intelligent cleaning devices placed on each floor in the management interface.

As an example of the present application, in order to facilitate parents to determine the location of their children, children usually wear smart wearable devices. In the case where parents accidentally separate from their children in a shopping mall, parents can use their mobile phones to locate their children's smart wearable devices to determine the floor where the children are located. For example, referring to Figure 7 (a), after parents open the application corresponding to the smart wearable device installed in the mobile phone, they can click the "indoor positioning" option in the search interface of the application; the mobile phone responds to the click operation of the "indoor positioning" option and sends a track acquisition request to the smart wearable device; when the smart wearable device receives the track acquisition request, it can send the track detection data collected within the specified time period to the parent's mobile phone; when the mobile phone receives the track detection data sent by the smart wearable device, it determines the floor where the smart wearable device is located based on the track detection data through the first classification model, and then, referring to Figure 7 (b), the mobile phone displays the current floor where the smart wearable device is located in the search interface, so that parents can know the current floor where the child is located.

As an example of the present application, after the car is parked in the parking space of the underground garage, if there are many floors in the underground garage and the driver did not pay attention to the floor where the car is parked during the parking process, then when looking for the parking space where the car is located later, the driver can use the mobile phone to perform indoor positioning of the car to determine the floor where the car is located. For example, the driver can open the car positioning application, see Figure (a) in Figure 8, the driver triggers the positioning operation of the car in the positioning interface of the car positioning application; in response to the positioning operation of the car, the mobile phone sends a trajectory acquisition request to the on-board device of the car; when the on-board device of the car receives the trajectory acquisition request, it can send the trajectory detection data within the specified time period before the car is turned off to the mobile phone. When the mobile phone receives the trajectory detection data sent by the on-board device of the car, based on the trajectory detection data, it is determined through the first classification model that the floor where the car is located is negative 2 floors, and then, see Figure (b) in Figure 8, the mobile phone displays the current floor of the car in the car positioning interface.

It should be noted that the embodiments of the present application are only described using the application scenarios shown in the above-mentioned Figures 5 to 8 as examples, and do not constitute a limitation on the embodiments of the present application.

Based on the execution entities and application scenarios provided in the above embodiments, the method for indoor positioning provided in the embodiments of the present application is introduced below. Please refer to Figure 9, which is a schematic flow chart of a method for indoor positioning according to an exemplary embodiment. As an example and not a limitation, the method is applied to an electronic device with a positioning function as an example, and the method may include some or all of the following contents:

Step 901: In response to a positioning operation on a positioning device, trajectory detection data of the positioning device is acquired.

It should be noted that the positioning device can be an electronic device that performs positioning, or can be another electronic device that is positioned by the electronic device. That is, the positioning device can locate its own location, or can be located by other electronic devices.

In a possible implementation, when the electronic device needs to locate itself, that is, when the electronic device is a positioning device, the electronic device responds to the positioning operation of the positioning device and obtains the trajectory detection data detected by the positioning device within a specified time period. Exemplarily, the process can refer to the application scenario shown in FIG5 above.

In another possible manner, when the electronic device and the positioning device are two independent devices and the electronic device positions the positioning device, in response to the positioning operation on the positioning device, the electronic device sends a track acquisition request to the positioning device; when the positioning device receives the track acquisition request, it acquires the track detection data detected within a specified time period and sends the track detection data to the electronic device; the electronic device receives the track detection data sent by the positioning device. Exemplarily, the process can refer to the application scenarios shown in FIG. 6, FIG. 7 or FIG. 8 above.

Step 902: Based on the trajectory detection data, determine the floor where the positioning device is located through a first classification model.

As an example, the electronic device can input the trajectory detection data into the first classification model, classify the trajectory detection data through the first classification model, and output the classification result of the trajectory detection data through the first classification model; and determine the classification result as the floor where the positioning device is located.

It should be noted that the first classification model is obtained by iteratively training the second classification model based on the first fingerprint library in advance, the first fingerprint library is obtained by data expansion of the second fingerprint library through the second classification model based on the fingerprint test set, and the second classification model is obtained by training the initial classification model based on the second fingerprint library. The second fingerprint library includes multiple sample trajectory detection data, and each of the multiple sample trajectory detection data has a floor label.

As an example, the first fingerprint library is obtained by performing data expansion on the second fingerprint library through the second classification model based on the fingerprint test set, and the operation of performing data expansion on the second fingerprint library through the second classification model based on the fingerprint test set includes: traversing the sample trajectory detection data in the fingerprint test set through the second classification model, and during the traversal process, adding the sample trajectory detection data whose classification result confidence is greater than or equal to the confidence threshold to the second fingerprint library. That is, the first fingerprint library is obtained by traversing the sample trajectory detection data in the fingerprint test set through the second classification model, and during the traversal process, adding the sample trajectory detection data whose classification result confidence is greater than or equal to the confidence threshold to the second fingerprint library.

It is worth noting that, in the process of testing the positioning performance of the second classification model based on the fingerprint test set, the data volume of the second fingerprint library can also be expanded, thereby ensuring the data volume of the first fingerprint library and improving the reliability of the first fingerprint library.

As an example, the second fingerprint library is obtained by the electronic device after performing data expansion processing on the initial fingerprint library based on the initial fingerprint library and the fingerprint test set, and the initial fingerprint library is obtained by the electronic device randomly selecting a second number of flat sample trajectory detection data from the acquired first number of sample trajectory detection data.

In this way, by randomly selecting the second number of flat-layer sample trajectory detection data, it is ensured that no cross-layer sample trajectory detection data will appear in the initial fingerprint library, thereby ensuring the reliability of the initial fingerprint library.

As an example, the fingerprint test set is a set of sample trajectory detection data remaining after the electronic device randomly selects the sample trajectory detection data from the first number of sample trajectory detection data.

It is worth noting that the fingerprint test set and the initial fingerprint library are derived from the same batch of sample trajectory detection data, thereby improving the convenience of constructing the initial fingerprint library and the fingerprint test set.

In some embodiments, the fingerprint test set may also be obtained in other ways, for example, the electronic device may also obtain other sample trajectory detection data in addition to the first number of sample trajectory detection data, and determine the set consisting of other sample trajectory detection data as the fingerprint test set. The other sample trajectory detection data is also sample trajectory detection data for the same indoor floor.

In some embodiments, the operation of the electronic device training the initial classification model based on the second fingerprint library to obtain the second classification model, the operation of performing data expansion on the second fingerprint library through the second classification model based on the fingerprint test set to obtain the first fingerprint library, and the operation of iteratively training the second classification model based on the first fingerprint library to obtain the first classification model can all refer to the method for constructing the fingerprint library shown in Figure 10 below.

In the embodiment of the present application, when indoor positioning is required, the electronic device can obtain the trajectory detection data of the positioning device, and then determine the floor where the positioning device is located through the first classification model based on the trajectory detection data. Since the first classification model is trained based on the first fingerprint library after data expansion, and the first fingerprint library is obtained by data expansion of the second fingerprint library based on the fingerprint test set, the amount of data in the first fingerprint library is relatively reliable, and thus the positioning accuracy of the first classification model is also improved.

Please refer to Figure 10, which is a schematic diagram of a method for constructing a fingerprint library according to an exemplary embodiment. As an example but not a limitation, the method is applied to an electronic device with a positioning function as an example for explanation, and the method may include some or all of the following contents:

Step 1001: Obtain a second fingerprint database and a fingerprint test set.

It should be noted that the second fingerprint library includes a plurality of sample trajectory detection data, and each of the plurality of sample trajectory detection data has a floor label.

In some embodiments, the operation of the electronic device acquiring the second fingerprint library and the fingerprint test set may refer to the operations of the following steps A1 to A4.

Step A1: the electronic device obtains a first quantity of sample trajectory detection data, where the first quantity of sample trajectory detection data at least includes flat layer sample trajectory detection data.

In some embodiments, in the process of building a fingerprint library, it is usually necessary to manually obtain sample trajectory detection data, that is, the collector carries an electronic device for data collection to collect sample trajectory detection data on each floor of the current indoor environment, and assigns a corresponding floor label to each sample trajectory detection data, and then sends the collected sample trajectory detection data carrying the floor label to the electronic device, so that the electronic device can obtain a first number of sample trajectory detection data.

Step A2: the electronic device randomly selects a second number of flat layer sample trajectory detection data from the first number of sample trajectory detection data to obtain an initial fingerprint library.

Since the data collector may collect cross-layer sample trajectory detection data in addition to flat-layer sample trajectory detection data during data collection, and the flat-layer sample trajectory detection data is data that is useful for positioning, the electronic device can randomly select a second number of flat-layer sample trajectory detection data from the first number of sample trajectories to obtain an initial fingerprint library. In addition, after subsequent model training, a fingerprint test set is usually set in order to verify the model training results. Therefore, the electronic device can determine the set consisting of the remaining sample trajectory detection data after random selection from the first number of sample trajectory detection data as the fingerprint test set.

It should be noted that the flat-floor sample trajectory detection data is the sample trajectory detection data collected by the electronic device for data collection on the same floor, and the cross-floor sample trajectory detection data is the sample trajectory detection data collected by the electronic device for data collection between different floors. In order to facilitate the understanding of different types of sample trajectory detection data, referring to FIG11, an embodiment of the present application provides a schematic diagram of different types of sample trajectory detection data, wherein the embodiment of the present application only uses the sample trajectory detection data shown in FIG11 as an example for illustration, and does not constitute a limitation on the embodiment of the present application.

It should be noted that the second number is a number preset according to demand, for example, the second number is 100 or 200.

In some cases, some sample trajectory detection data include a large number of trajectory points, which may cause trouble for the calculation amount of subsequent positioning. Therefore, the electronic device randomly selects a second number of flat layer sample trajectory detection data from the first number of sample trajectory detection data. Before obtaining the initial fingerprint library, the sample trajectory detection data including a large number of trajectory points may be processed to reduce the number of trajectory points in the sample trajectory detection data.

In some embodiments, when the first number of sample trajectory detection data includes sample trajectory detection data whose number of trajectory points is greater than the number threshold, the sample trajectory detection data whose number of trajectory points is greater than the number threshold is segmented to obtain a third number of sample trajectory detection data, wherein the number of trajectory points of each sample trajectory detection data in the third number of sample trajectory detection data is less than or equal to the number threshold. In this way, the electronic device can randomly select a second number of sample trajectory detection data from the third number of sample trajectory detection data to obtain an initial fingerprint library.

When the number of trajectory points included in the sample trajectory detection data is greater than the quantity threshold, it means that subsequent positioning operations based on the sample trajectory detection data will cause computational troubles. Therefore, the electronic device can segment the sample trajectory detection data whose number of trajectory points is greater than the quantity threshold.

It should be noted that the quantity threshold can be pre-set according to requirements, for example, the quantity threshold can be 15, 20 or 30, etc.

It is worth noting that by segmenting the longer sample trajectory detection data, the number of trajectory points included in the longer sample trajectory detection data is reduced, thereby reducing the amount of subsequent calculations based on the sample trajectory detection data.

As an example, in a case where the first number of sample trajectory detection data includes sample trajectory detection data whose number of trajectory points is greater than a quantity threshold, the operation of the electronic device performing segmentation processing on the sample trajectory detection data whose number of trajectory points is greater than the quantity threshold includes: in a case where the first number of sample trajectory detection data includes sample trajectory detection data whose number of trajectory points is greater than the quantity threshold, segmenting the sample trajectory detection data whose number of trajectory points is greater than the quantity threshold with a segmentation value as an interval to obtain a fourth number of sample trajectory detection data, where the segmentation value is a value less than or equal to the quantity threshold; in a case where the fourth number of sample trajectory detection data includes sample trajectory detection data whose number of trajectory points is less than the segmentation value, deleting the sample trajectory detection data whose number of trajectory points is less than the segmentation value to obtain a third number of sample trajectory detection data.

It should be noted that the division value can also be pre-set according to needs, and the division value can be the same as the quantity threshold or less than the quantity threshold, and the embodiment of the present application does not impose specific restrictions on this.

Because when the number of trajectory points included in the sample trajectory detection data is too small, the sample trajectory detection data may affect the accuracy of model training, and therefore, the electronic device may delete the sample trajectory detection data whose number of trajectory points is less than the division value.

It is worth noting that by dividing the sample trajectory detection data whose number of trajectory points is greater than the quantity threshold into intervals, not only can the number of sample trajectory detection data be expanded, but also the amount of calculation in subsequent model training can be reduced and the accuracy of model training can be improved.

Since the flat layer sample trajectory data is the sample trajectory data for model training, when the electronic device performs segmentation processing on the sample trajectory detection data whose number of trajectory points is greater than the number threshold, it can perform segmentation processing on the flat layer sample trajectory detection data whose number of trajectory points is greater than the number threshold.

Step A3: The electronic device determines a set consisting of sample trajectory detection data remaining after randomly selecting the first number of sample trajectory detection data as a fingerprint test set.

In some embodiments, the electronic device may not only determine the set consisting of the sample trajectory detection data remaining after random selection from the first number of sample trajectory detection data as the fingerprint test set, but may also determine it in other ways, for example, the electronic device may also obtain other sample trajectory detection data other than the first number of sample trajectory detection data, and determine the set consisting of other sample trajectory detection data as the fingerprint test set. The other sample trajectory detection data is also the sample trajectory detection data for the same indoor floor.

Step A4: The electronic device performs data expansion processing on the initial fingerprint library based on the initial fingerprint library and the fingerprint test set to obtain a second fingerprint library.

Since the number of sample trajectory detection data in the initial fingerprint library is small, it affects both the accuracy of model training and the accuracy of subsequent positioning. Therefore, in order to improve the accuracy of subsequent model training and positioning, the electronic device can perform data expansion processing on the initial fingerprint library based on the initial fingerprint library and the fingerprint test set.

It is worth noting that the initial fingerprint library is expanded by using the fingerprint test set, thereby ensuring the data volume of the sample trajectory detection data in the second fingerprint library and improving the reliability of the second fingerprint library.

In a possible implementation, the electronic device performs data expansion processing on the initial fingerprint library based on the initial fingerprint library and the fingerprint test set, including: traversing each sample trajectory detection data in the fingerprint test set in turn; determining the Euclidean distance and the Jaccard distance between the second sample trajectory detection data and each sample trajectory detection data in the initial fingerprint library, the second sample trajectory detection data being the currently traversed sample trajectory detection data; in the case where there is at least one third sample trajectory detection data in the initial fingerprint library, determining the third sample trajectory detection data with the greatest similarity to the second sample trajectory detection data from the at least one third sample trajectory detection data, the third sample trajectory detection data referring to the sample trajectory detection data having a Euclidean distance with the second sample trajectory detection data less than a first distance threshold and a Jaccard distance with the second sample trajectory detection data less than a second distance threshold; updating the floor label of the second sample trajectory detection data to the determined floor label of the third sample trajectory detection data; and adding the second sample trajectory detection data with the updated label to the initial fingerprint library.

Since the Euclidean distance and the Jaccard distance can describe the similarity between two data, and the smaller the distance, the greater the similarity, the electronic device can determine the Euclidean distance and the Jaccard distance between the second sample trajectory detection data and each sample trajectory detection data in the initial fingerprint library.

In some embodiments, the electronic device adds the second sample trajectory detection data with the updated label to the initial fingerprint library, and deletes the second sample trajectory detection data from the fingerprint test set.

In some embodiments, when there is no at least one third sample trajectory detection data, the electronic device may end the expansion of the initial fingerprint library.

It is worth noting that by determining the third sample trajectory detection data having the greatest similarity to the second sample trajectory detection data from at least one third sample trajectory detection data, the rationality of expanding the initial fingerprint library is ensured.

In order to facilitate the determination of the Euclidean distance and Jaccard distance between the second sample trajectory detection data and each sample trajectory detection data in the initial fingerprint library, for each trajectory point in each sample trajectory detection data in the fingerprint test set and each trajectory point in each sample trajectory detection data in the initial fingerprint library, the electronic device can be represented by m wifi signal strengths, where m is the number of wireless access points (AP) installed in the current indoor environment. For example, m can be 8000, 10000 or 100, etc.

Exemplarily, the initial fingerprint library F can be expressed by the following first formula (1), wherein, in the following first formula (1), the initial fingerprint library F includes N sample trajectory detection data; any sample trajectory detection data _Bi includes x trajectory points, and the number of trajectory points included in different sample trajectory detection data may be different, so the value of x is a variable value; each trajectory point _Ci is represented by m WiFi signal strengths, and AP ₁ represents the WiFi signal strength of the WiFi signal corresponding to AP ₁ .

Exemplarily, the fingerprint test set T can be expressed by the following second formula (2), wherein in the following second formula (2), the fingerprint test set T includes P sample trajectory detection data, and any sample trajectory detection data _Ai includes x trajectory points. Similarly, the value of x can be a variable value; each trajectory point _Di is represented by m WiFi signal strengths, and AP ₁ represents the WiFi signal strength of the WiFi signal corresponding to AP ₁ .

In some embodiments, the electronic device may determine the Euclidean distance between the second sample trajectory detection data and each sample trajectory detection data in the initial fingerprint library by using the following third formula (3).

d _{Ai_F} =min _i∈(1,N) |A _j -B _i | (3)

It should be noted that in the above third formula (3), d _{Aj_F} is the Euclidean distance between the second sample trajectory detection data and any sample trajectory detection data in the initial fingerprint library, A _j is the second sample trajectory detection data, and _Bi is any sample trajectory detection data in the initial fingerprint library.

In some embodiments, the electronic device may determine the Jaccard distance between the second sample trajectory detection data and each sample trajectory detection data in the initial fingerprint library by using the following fourth formula (4).

It should be noted that in the fourth formula (4), is the Jaccard distance between the second sample trajectory detection data and any sample trajectory detection data in the initial fingerprint library, A _j is the second sample trajectory detection data, and _Bi is any sample trajectory detection data in the initial fingerprint library.

Since the electronic device can determine the Euclidean distance and the Jaccard distance between the second sample trajectory detection data and each sample trajectory detection data in the initial fingerprint library, the electronic device needs to determine the sample trajectory detection data that satisfies both the Euclidean distance requirement and the Jaccard distance requirement, that is, the electronic device needs to determine at least one third sample trajectory detection data of the sample trajectory detection data whose Euclidean distance with the second sample trajectory detection data is less than the first distance threshold and whose Jaccard distance with the second sample trajectory detection data is less than the second distance threshold.

It should be noted that both the first distance threshold and the second distance threshold can be pre-set according to requirements, and the first distance threshold and the second distance threshold can be fixed thresholds or dynamic thresholds.

Since in the process of constructing the fingerprint library, the electronic device selects a fixed number of sample trajectory detection data from the sample trajectory detection data corresponding to each floor and adds them to the fingerprint library, and the sample trajectory detection data collected manually usually has limitations, for example, the sample trajectory detection data collected manually presents the phenomenon of uneven data distribution, that is, the number of sample trajectory detection data corresponding to some floors is large, while the number of sample trajectory detection data corresponding to some floors is small, and even the number of sample trajectory detection data corresponding to some floors may not reach the required number. Therefore, the first distance threshold and the second distance threshold can also be dynamic thresholds, and the first distance threshold and the second distance threshold can both be determined according to the number of sample trajectory detection data corresponding to each floor in the initial fingerprint library. Among them, when the number of sample trajectory detection data corresponding to any floor is small, in order to perform a large amount of data expansion on the sample trajectory detection data corresponding to the floor, the first distance threshold and the second distance threshold corresponding to the floor are both relatively large; when the number of sample trajectory detection data corresponding to any floor is large, the first distance threshold and the second distance threshold corresponding to the floor are both relatively small.

Exemplarily, a correspondence between the quantity, the first distance threshold and the second distance threshold is set in the electronic device. Therefore, the electronic device can determine the first distance threshold and the second distance threshold corresponding to each sample trajectory detection data from the correspondence according to the quantity of sample trajectory detection data corresponding to each floor in the initial fingerprint database.

Since the first distance threshold and the second distance threshold are dynamic thresholds, after determining the Euclidean distance and Jaccard distance between the second sample trajectory detection data and each sample trajectory detection data in the initial fingerprint library, the electronic device can determine the first distance threshold and the second distance threshold corresponding to each sample trajectory detection data according to the number of sample trajectory detection data corresponding to each floor in the initial fingerprint library, and then determine whether there is at least one third sample trajectory detection data according to the first distance threshold and the second distance threshold corresponding to each sample trajectory detection data; in the case of at least one third sample trajectory detection data, determine the third sample trajectory detection data with the greatest similarity from the at least one third sample trajectory detection data, update the floor label of the second sample trajectory detection data to the floor label of the determined third sample trajectory detection data, and add the second sample trajectory detection data with the updated label to the initial fingerprint library, and at the same time, delete the second sample trajectory detection data from the fingerprint test set. In the case of the absence of at least one third sample trajectory detection data, the expansion of the initial fingerprint library is terminated. Exemplarily, the process can be represented by the schematic diagram shown in FIG. 12 below.

It is worth noting that by setting the first distance threshold and the second distance threshold, after the third sample trajectory detection data after the label update is added to the initial fingerprint library, the sample trajectory detection data corresponding to each floor in the second fingerprint library is made more balanced.

In some embodiments, when there is at least one third sample trajectory detection data in the initial fingerprint library, the operation of the electronic device determining the third sample trajectory detection data having the greatest similarity to the second sample trajectory detection data from the at least one third sample trajectory detection data includes: when there is at least one third sample trajectory detection data in the initial fingerprint library, obtaining the third sample trajectory detection data having the smallest Euclidean distance with the second sample trajectory data from the at least one third sample trajectory detection data; or, when there is at least one third sample trajectory detection data in the initial fingerprint library, obtaining the third sample trajectory detection data having the smallest Jaccard distance with the second sample trajectory data from the at least one third sample trajectory detection data.

Since the smaller the Euclidean distance, the greater the similarity, therefore, in the case where there is at least one third sample trajectory detection data in the initial fingerprint library, the electronic device can select the third sample trajectory detection data with the smallest Euclidean distance from the second sample trajectory data from at least one third sample trajectory, and the selected third sample trajectory detection data is the sample trajectory detection data with the greatest similarity to the second sample trajectory detection data. Of course, since the Jaccard distance can also measure the similarity between two data, and the smaller the Jaccard distance, the greater the similarity, therefore, in the case where there is at least one third sample trajectory detection data in the initial fingerprint library, the electronic device can also select the third sample trajectory detection data with the smallest Jaccard distance from the second sample trajectory data from at least one third sample trajectory, and the selected third sample trajectory detection data is the sample trajectory detection data with the greatest similarity to the second sample trajectory detection data.

It is worth noting that, since both the Euclidean distance and the Jaccard distance can describe the similarity between two data, by using the Euclidean distance or the Jaccard distance, the third sample trajectory detection data having the greatest similarity with the second sample trajectory detection data is selected from at least one third sample trajectory detection data, so that the selected third sample trajectory detection data can be more accurate.

In some embodiments, since both the Euclidean distance and the Jaccard distance can describe the similarity between two data, when the Euclidean distance between the two data is the smallest, the Jaccard distance between the two data may also be the smallest. Therefore, the electronic device can not only determine the third sample trajectory detection data with the greatest similarity to the second sample trajectory detection data from at least one third sample trajectory detection data in the above manner, but can also determine it in other ways. For example, when there is at least one third sample trajectory detection data in the initial fingerprint library, the third sample trajectory detection data with the smallest Euclidean distance to the second sample trajectory data and the smallest Jaccard distance to the second sample trajectory data is obtained from at least one third sample trajectory detection data.

In another possible implementation, the electronic device performs data expansion processing on the initial fingerprint library based on the initial fingerprint library and the fingerprint test set, including: traversing each sample trajectory detection data in the fingerprint test set in turn; determining the similarity between the second sample trajectory detection data and each sample trajectory detection data in the initial fingerprint library to obtain multiple similarities, and the second sample trajectory detection data is the sample trajectory detection data currently traversed; when the maximum similarity among the multiple similarities is greater than or equal to the similarity threshold, updating the floor label of the second sample trajectory detection data to the floor label of the sample trajectory detection data corresponding to the maximum similarity; and adding the second sample trajectory detection data with the updated label to the initial fingerprint library.

As an example, the electronic device may represent the similarity between the second sample trajectory detection data and each sample trajectory detection data in the initial fingerprint library by any one of the Euclidean distance, Jaccard distance, cosine distance, Manhattan distance, Chebyshev distance, etc. between the second sample trajectory detection data and each sample trajectory detection data in the initial fingerprint library.

It should be noted that the similarity threshold can be set in advance according to the similarity algorithm selected by the electronic device.

It is worth noting that the data expansion processing is performed on the initial fingerprint library in different ways, thereby increasing the richness of the determined data expansion methods.

Step 1002: iteratively train the initial classification model based on the second fingerprint library to obtain a second classification model.

It should be noted that the electronic device iteratively trains the initial classification model based on the second fingerprint library, and the operation of obtaining the second classification model may include a variety of different methods. For example, the electronic device can input each sample trajectory detection data in the second fingerprint library into the initial classification model, and train the initial classification model based on the WiFi signal strength corresponding to each trajectory point in each sample trajectory detection data, and then continue to iterate the training according to the training results of each trajectory point in each sample trajectory detection data to obtain the second classification model.

Since each trajectory point in each sample trajectory detection data can be represented by m WiFi signal strengths, and the corresponding m WiFi signal strengths are different depending on the position of the trajectory point, but each trajectory point belongs to the same floor, that is, the floor label corresponding to each trajectory point in each sample trajectory detection data is the same. In the case where any sample trajectory detection data includes multiple trajectory points, the sample trajectory detection data corresponds to multiple groups of m WiFi signal strengths, and the floor labels corresponding to the multiple groups of m WiFi signal strengths are also the same. Therefore, in order to fully train the initial classification model, the electronic device can train the initial classification model based on the WiFi signal strength corresponding to each trajectory point in each sample trajectory detection data. In some embodiments, the electronic device also includes multiple ways to train the initial classification model based on the WiFi signal strength corresponding to each trajectory point in each sample trajectory detection data, and the embodiments of the present application will not be repeated one by one.

It is worth noting that the electronic device trains the initial classification model based on the WiFi signal strength corresponding to each trajectory point in each sample trajectory detection data, so that the initial classification model can recognize the characteristics of the WiFi signal strength corresponding to the trajectory points on each floor, and then subsequently identify trajectory points that do not belong to the same floor.

In some embodiments, before the electronic device iteratively trains the initial classification model based on the second fingerprint library to obtain the second classification model, it can also perform dimensionality reduction processing on each sample trajectory detection data in the second fingerprint library.

Exemplarily, the electronic device may perform dimensionality reduction processing on each sample trajectory detection data in the second fingerprint library through algorithms such as principal component analysis (PCA) or singular value decomposition (SVD), so that the dimension of each sample trajectory detection data after dimensionality reduction processing is the target dimension.

It should be noted that the target dimension is a dimension pre-set according to demand, for example, the target dimension is 64 dimensions.

Step 1003: traverse and process the sample trajectory detection data in the fingerprint test set through the second classification model.

In order to determine the positioning performance of the second classification model, the electronic device may test the second classification model based on the fingerprint test set. That is, the electronic device may traverse and process the sample trajectory detection data in the fingerprint test set through the second classification model.

In some embodiments, the electronic device may input each sample trajectory detection data in the fingerprint test set into the second classification model, and perform classification processing on each sample trajectory detection data in the fingerprint test set through the second classification model to complete the traversal processing of the sample trajectory detection data in the fingerprint test set.

In some embodiments, before the electronic device traverses the sample trajectory detection data in the fingerprint test set through the second classification model, it can also perform dimensionality reduction processing on each sample trajectory detection data in the fingerprint test set. And the electronic device can perform dimensionality reduction processing on each sample trajectory detection data in the second fingerprint library at the same time as the dimensionality reduction processing on each sample trajectory detection data in the fingerprint test set. Of course, it is also possible to first perform dimensionality reduction processing on each sample trajectory detection data in the second fingerprint library, or first perform dimensionality reduction processing on each sample trajectory detection data in the fingerprint test set. The embodiment of the present application does not specifically limit the order of dimensionality reduction processing.

Exemplarily, the electronic device may perform dimensionality reduction processing on each sample trajectory detection data in the fingerprint test set by using algorithms such as PCA or SVD, so that the dimension of each sample trajectory detection data after the dimensionality reduction processing is the target dimension.

Step 1004: during the traversal process, the sample trajectory detection data whose classification result confidence is greater than or equal to the confidence threshold is added to the second fingerprint library.

In order to improve the reliability of the fingerprint library, the electronic device can also add sample trajectory detection data whose classification result confidence is greater than or equal to the confidence threshold to the second fingerprint library during the process of testing the positioning performance of the second classification model based on the fingerprint test set test.

In some embodiments, the classification result includes a floor label and a probability value corresponding to the floor label; during the traversal process, the electronic device adds the sample trajectory detection data whose confidence in the classification result is greater than or equal to the confidence threshold to the second fingerprint library, including: during the traversal process, according to the probability value in the classification result corresponding to the first sample trajectory detection data, determining the confidence of the classification result corresponding to the first sample trajectory detection data, the first sample trajectory detection data being any sample trajectory detection data in the currently traversed fingerprint test set; when the confidence in the classification result corresponding to the first sample trajectory detection data is greater than or equal to the confidence threshold, updating the floor label of the first sample trajectory detection data to the floor label in the classification result corresponding to the first sample trajectory detection data; and adding the first sample trajectory detection data with the updated label to the second fingerprint library.

Since the classification result may include multiple floor labels and probability values corresponding to each of the multiple floor labels after the electronic device classifies the first sample trajectory detection data through the second classification model, the electronic device can obtain the maximum probability value from the probability values corresponding to each of the multiple floor labels, and determine the maximum probability value as the confidence of the classification result.

In some embodiments, the electronic device may also determine the confidence of the classification result in other ways, for example, the electronic device may also obtain the maximum probability value from the probability values corresponding to each floor label in the multiple floor labels; multiply the maximum probability value by a preset coefficient to obtain a product result; and determine the product result as the confidence of the classification result. Alternatively, the electronic device may select at least one probability value greater than the preset probability value from the probability values corresponding to each floor label in the multiple floor labels; determine the average value of the at least one probability value obtained; and determine the average value as the confidence of the classification result.

It should be noted that the confidence threshold can be set in advance according to the requirements, for example, the confidence threshold can be 0.7, 0.8 or 0.9, etc. The preset probability value can be set in advance, for example, the preset probability value can be 40%, 35% or 30%, etc.

In some embodiments, the electronic device may delete the first sample trajectory detection data from the fingerprint test set while adding the first sample trajectory detection data after the label is updated to the second fingerprint library.

It is worth noting that, in the process of testing the second classification model based on the fingerprint test set, the data of the second fingerprint library can also be expanded, so that the model training and fingerprint library construction can complement each other.

Step 1005: When the traversal is completed, the second classification model is iteratively trained based on the first fingerprint library to obtain the first classification model.

It should be noted that the first fingerprint library is obtained after the second fingerprint library is expanded during the traversal process, that is, the first fingerprint library is obtained after the first sample trajectory detection data whose classification result confidence is greater than or equal to the confidence threshold in the fingerprint test set is updated with a label and added to the second fingerprint library during the traversal process according to the operation of the above step 1004. The first classification model can determine the floor where the positioning device is located based on the trajectory detection data of any positioning device.

In some embodiments, after the electronic device traverses and processes the sample trajectory detection data in the fingerprint test set through the second classification model, it can also obtain the correct amount of the sample trajectory detection data in the fingerprint test set classified by the second classification model, as well as the added amount of sample trajectory detection data added to the second fingerprint library after traversing and processing the sample trajectory detection data in the fingerprint test set through the second classification model; the correct amount is divided by the total number of samples in the fingerprint test set to obtain the classification accuracy, and the total number of samples in the fingerprint test set is the number of sample trajectory detection data included in the fingerprint test set before the sample trajectory detection data in the fingerprint test set is added to the second fingerprint library through the second classification model; the added amount is divided by the total number of samples to obtain the trajectory utilization, and the classification accuracy and the trajectory utilization are used to evaluate the positioning performance of the first classification model.

It is worth noting that by determining the classification accuracy of the second classification model and the trajectory utilization rate, the positioning performance of the second classification model can be evaluated using specific data, thereby improving the accuracy of evaluating the positioning performance of the second classification model.

After the sample trajectory data of the fingerprint test set is traversed and processed by the second classification model, the electronic device will iteratively train the second classification model based on the first fingerprint library to obtain the first classification model, that is, the first classification model is iteratively trained on the basis of the second classification model, and the positioning performance of the first classification model is very likely to be higher than or equal to the positioning performance of the second classification model. Therefore, the classification accuracy and trajectory utilization of the second classification model can be used to evaluate the positioning performance of the second classification model, and can also be used to evaluate the positioning performance of the first classification model.

It should be noted that the evaluation of the positioning performance of the first classification model can be performed manually or by an electronic device based on the classification accuracy and trajectory utilization.

As an example, the electronic device may compare the classification accuracy with the accuracy threshold, and compare the trajectory utilization with the utilization threshold; when the classification accuracy is greater than or equal to the accuracy threshold, and the trajectory utilization is greater than or equal to the utilization threshold, the positioning performance of the first classification model is determined to be the first-level performance; when the classification accuracy is less than the accuracy threshold, or the trajectory utilization is less than the utilization threshold, the positioning performance of the first classification model is determined to be the second-level performance; when the classification accuracy is less than the accuracy threshold, and the trajectory utilization is less than the utilization threshold, the positioning performance of the first classification model is determined to be the third-level performance. Among them, the positioning accuracy of the classification model with the first-level performance is greater than the positioning accuracy of the classification model with the second-level performance, and the positioning accuracy of the classification model with the second-level performance is greater than the positioning accuracy of the classification model with the third-level performance.

It should be noted that both the accuracy threshold and the utilization threshold can be preset. For example, the accuracy threshold can be 90% or 80%, and the utilization threshold can be 60% or 70%.

In some embodiments, the electronic device iteratively trains the second classification model based on the first fingerprint library, and after obtaining the first classification model, it can continue to traverse and process the remaining sample trajectory data in the fingerprint test set through the first classification model. The operation of traversing and processing the remaining sample trajectory data in the fingerprint test set through the first classification model is the same or similar to the operation of traversing and processing the sample trajectory data in the fingerprint test set through the second classification model, and the embodiments of the present application will not be described one by one.

In an embodiment of the present application, when it is necessary to perform indoor positioning of the positioning device, the trajectory detection data of the positioning device can be obtained, and then the floor where the positioning device is located can be determined through the first classification model based on the trajectory detection data. Since the first classification model is trained based on the first fingerprint library after data expansion, and the first fingerprint library is obtained by data expansion of the second fingerprint library based on the fingerprint test set, the amount of data in the first fingerprint library is more reliable, and the positioning accuracy of the first classification model is also improved. That is, the electronic device in the embodiment of the present application not only expands the amount of data in the fingerprint library and improves the reliability of the fingerprint library, but also improves the positioning accuracy of the classification model.

Please refer to FIG. 13, which is a schematic diagram of a method for constructing a fingerprint library according to another exemplary embodiment. As an example but not a limitation, the method is described here by applying it to an electronic device with a positioning function. The method may include some or all of the following contents:

Step 1301: Obtain a first amount of manually collected sample trajectory detection data.

Step 1302: Determine whether the first number of sample trajectory detection data includes flat layer sample trajectory detection data whose number of trajectory points is greater than the quantity threshold; if yes, perform the operation of the following step 1303; if no, perform the operation of the following step 1304.

Step 1303: When the first number of sample trajectory detection data includes evaluation sample trajectory detection data whose number of trajectory points is greater than the number threshold, segment the leveling sample trajectory detection data whose number of trajectory points is greater than the number threshold to obtain a third number of sample trajectory detection data.

Step 1304: Represent each trajectory point in each sample trajectory detection data by m WiFi signal strengths.

Step 1305: randomly select a second number of flat layer sample trajectory detection data to obtain an initial fingerprint library.

It should be noted that, when the first number of sample trajectory detection data does not include flat sample trajectory detection data whose number of trajectory points is greater than the number threshold, the electronic device randomly selects a second number of flat sample trajectory detection data from the first number of sample trajectory detection data to obtain an initial fingerprint library. When the first number of sample trajectory detection data includes flat sample trajectory detection data whose number of trajectory points is greater than the number threshold, the electronic device randomly selects a second number of flat sample trajectory detection data from the third number of sample trajectory detection data to obtain an initial fingerprint library.

Step 1306: Determine the set consisting of the sample trajectory detection data remaining after the random selection as the fingerprint test set.

Step 1307: Based on the initial fingerprint library and the fingerprint test set, perform data expansion processing on the initial fingerprint library to obtain a second fingerprint library.

Step 1308: Iteratively train the initial classification model based on the second fingerprint library to obtain a second classification model.

Step 1309: Perform traversal processing on the sample trajectory detection data in the fingerprint test set through the second classification model.

Step 1310: During the traversal process, determine whether there is sample trajectory detection data whose classification result confidence is greater than or equal to the confidence threshold. If so, perform the operation of the following step 1311; if not, perform the operation of the following step 1313.

Step 1311: adding the sample trajectory detection data whose classification result confidence is greater than or equal to the confidence threshold to the second fingerprint library.

Step 1312: iteratively train the second classification model based on the first fingerprint library to obtain the first classification model.

It should be noted that, when the first classification model is obtained, the electronic device may continue to perform the operation of step 1313 described below.

Step 1313: Determine the classification accuracy and trajectory utilization of the second classification model.

Step 1314: End model training.

It should be noted that the operations of the above steps 1301 to 1314 can refer to the operations of the above steps 1001 to 1005, and the embodiments of the present application will not describe them one by one.

In an embodiment of the present application, when it is necessary to perform indoor positioning of the positioning device, the trajectory detection data of the positioning device can be obtained, and then the floor where the positioning device is located can be determined through the first classification model based on the trajectory detection data. Since the first classification model is trained based on the first fingerprint library after data expansion, and the first fingerprint library is obtained by data expansion of the second fingerprint library based on the fingerprint test set, the amount of data in the first fingerprint library is more reliable, and the positioning accuracy of the first classification model is also improved. That is, the electronic device in the embodiment of the present application not only expands the amount of data in the fingerprint library and improves the reliability of the fingerprint library, but also improves the positioning accuracy of the classification model.

FIG14 is a schematic diagram of the structure of an indoor positioning device provided by an embodiment of the present application, which can be implemented by software, hardware or a combination of both to form part or all of a computer device, which can be the computer device shown in FIG3. Referring to FIG14, the device includes: an acquisition module 1401 and a determination module 1402.

An acquisition module 1401 is used for acquiring trajectory detection data of the positioning device in response to a positioning operation on the positioning device;

A determination module 1402 is used to determine the floor where the positioning device is located through a first classification model based on the trajectory detection data;

The first classification model is obtained by iteratively training the second classification model based on the first fingerprint library in advance, the first fingerprint library is obtained by expanding the data of the second fingerprint library through the second classification model based on the fingerprint test set, the second classification model is obtained by training the initial classification model based on the second fingerprint library, and the second fingerprint library includes multiple sample trajectory detection data, and each of the multiple sample trajectory detection data has a floor label.

As an example of the present application, the first fingerprint library is obtained by traversing the sample trajectory detection data in the fingerprint test set through the second classification model. During the traversal process, the sample trajectory detection data whose classification result confidence is greater than or equal to the confidence threshold is added to the second fingerprint library.

As an example of the present application, the second fingerprint library is obtained after data expansion processing is performed on the initial fingerprint library based on the initial fingerprint library and the fingerprint test set, and the initial fingerprint library is obtained by randomly selecting a second number of flat layer sample trajectory detection data from the acquired first number of sample trajectory detection data.

As an example of the present application, the fingerprint test set is a set consisting of remaining sample trajectory detection data after randomly selecting from the first number of sample trajectory detection data.

In the embodiment of the present application, when indoor positioning is required, the trajectory detection data of the positioning device can be obtained, and then the floor where the positioning device is located can be determined by the first classification model based on the trajectory detection data. Since the first classification model is trained based on the first fingerprint library after data expansion, and the first fingerprint library is obtained by data expansion of the second fingerprint library based on the fingerprint test set, the amount of data in the first fingerprint library is relatively reliable, and thus the positioning accuracy of the first classification model is also improved.

It should be noted that: the indoor positioning device provided in the above embodiment only uses the division of the above-mentioned functional modules as an example when performing indoor positioning. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

The functional units and modules in the above embodiments may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit, and the above integrated units may be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the protection scope of the embodiments of the present application.

The indoor positioning device and the indoor positioning method embodiments provided in the above embodiments belong to the same concept. The specific working process of the units and modules in the above embodiments and the technical effects brought about can be found in the method embodiment part and will not be repeated here.

FIG15 is a schematic diagram of the structure of a fingerprint library construction device provided in an embodiment of the present application, and the device can be implemented by software, hardware or a combination of both to become part or all of a computer device, and the computer device can be the computer device shown in FIG3. Referring to FIG15, the device includes: a first acquisition module 1501, a first training module 1502, a traversal module 1503, an addition module 1504 and a second training module 1505.

The first acquisition module 1501 is used to acquire a second fingerprint library and a fingerprint test set, wherein the second fingerprint library includes a plurality of sample trajectory detection data, and each of the plurality of sample trajectory detection data has a floor label;

A first training module 1502, configured to iteratively train the initial classification model based on the second fingerprint library to obtain a second classification model;

A traversal module 1503, configured to perform traversal processing on the sample trajectory detection data in the fingerprint test set by using the second classification model;

An adding module 1504 is used to add the sample trajectory detection data whose classification result confidence is greater than or equal to the confidence threshold to the second fingerprint library during the traversal process;

The second training module 1505 is used to iteratively train the second classification model based on the first fingerprint library to obtain a first classification model when the traversal is completed. The first fingerprint library is obtained by expanding the second fingerprint library during the traversal process. The first classification model can determine the floor where the positioning device is located based on the trajectory detection data of any positioning device.

As an example of the present application, the classification result includes a floor label and a probability value corresponding to the floor label;

The adding module 1504 is used for:

During the traversal process, according to the probability value in the classification result corresponding to the first sample trajectory detection data, the confidence of the classification result corresponding to the first sample trajectory detection data is determined, and the first sample trajectory detection data is any sample trajectory detection data in the fingerprint test set currently traversed;

When the confidence of the classification result corresponding to the first sample trajectory detection data is greater than or equal to the confidence threshold, updating the floor label of the first sample trajectory detection data to the floor label in the classification result corresponding to the first sample trajectory detection data;

The first sample trajectory detection data after label update is added to the second fingerprint library.

As an example of the present application, the first acquisition module 1501 is used to:

Acquire a first number of sample trajectory detection data, where the first number of sample trajectory detection data at least includes level layer sample trajectory detection data;

Randomly selecting a second number of flat layer sample trajectory detection data from the first number of sample trajectory detection data to obtain an initial fingerprint library;

Determine a set consisting of remaining sample trajectory detection data after randomly selecting the first number of sample trajectory detection data as the fingerprint test set;

Based on the initial fingerprint library and the fingerprint test set, data expansion processing is performed on the initial fingerprint library to obtain the second fingerprint library.

As an example of the present application, the first acquisition module 1501 is used to:

Traversing each sample trajectory detection data in the fingerprint test set in turn;

Determine the Euclidean distance and the Jaccard distance between the second sample trajectory detection data and each sample trajectory detection data in the initial fingerprint library, wherein the second sample trajectory detection data is the currently traversed sample trajectory detection data;

In the case where there is at least one third sample trajectory detection data in the initial fingerprint library, determining the third sample trajectory detection data having the greatest similarity to the second sample trajectory detection data from the at least one third sample trajectory detection data, wherein the third sample trajectory detection data refers to the sample trajectory detection data having a Euclidean distance with the second sample trajectory detection data less than a first distance threshold and a Jaccard distance with the second sample trajectory detection data less than a second distance threshold;

Updating the floor label of the second sample trajectory detection data to the determined floor label of the third sample trajectory detection data;

The second sample trajectory detection data after the label is updated is added to the initial fingerprint library.

As an example of the present application, the first acquisition module 1501 is used to:

In the case where there is at least one third sample trajectory detection data in the initial fingerprint library, obtaining the third sample trajectory detection data having the smallest Euclidean distance with the second sample trajectory data from the at least one third sample trajectory detection data; or,

In the case that at least one third sample trajectory detection data exists in the initial fingerprint library, the third sample trajectory detection data having the smallest Jaccard distance with the second sample trajectory data is obtained from the at least one third sample trajectory detection data.

As an example of the present application, the first acquisition module 1501 is used to:

Traversing each sample trajectory detection data in the fingerprint test set in turn;

Determine the similarity between the second sample trajectory detection data and each sample trajectory detection data in the initial fingerprint library to obtain multiple similarities, wherein the second sample trajectory detection data is the currently traversed sample trajectory detection data;

When the maximum similarity among the multiple similarities is greater than or equal to the similarity threshold, updating the floor label of the second sample trajectory detection data to the floor label of the sample trajectory detection data corresponding to the maximum similarity;

The second sample trajectory detection data after the label is updated is added to the initial fingerprint library.

As an example of the present application, the first acquisition module 1501 is further used for:

In a case where the first number of sample trajectory detection data includes sample trajectory detection data whose number of trajectory points is greater than a number threshold, segmenting the sample trajectory detection data whose number of trajectory points is greater than the number threshold to obtain a third number of sample trajectory detection data, wherein the number of trajectory points of each sample trajectory detection data in the third number of sample trajectory detection data is less than or equal to the number threshold;

The second number of flat layer sample trajectory detection data is randomly selected from the third number of sample trajectory detection data to obtain the initial fingerprint library.

As an example of the present application, the first acquisition module 1501 is used to:

In a case where the first number of sample trajectory detection data includes sample trajectory detection data whose number of trajectory points is greater than the number threshold, segmenting the sample trajectory detection data whose number of trajectory points is greater than the number threshold at intervals of a segmentation value to obtain a fourth number of sample trajectory detection data, wherein the segmentation value is a value less than or equal to the number threshold;

In the case that the fourth amount of sample trajectory detection data includes sample trajectory detection data whose number of trajectory points is smaller than the division value, the sample trajectory detection data whose number of trajectory points is smaller than the division value is deleted.

As an example of the present application, the device further includes:

A second acquisition module is used to obtain the correct amount of sample trajectory detection data in the fingerprint test set classified by the second classification model, and the amount of sample trajectory detection data added to the second fingerprint library after traversing the sample trajectory detection data in the fingerprint test set through the second classification model;

a first calculation module, configured to obtain the classification accuracy by dividing the correct amount by the total number of samples in the fingerprint test set, where the total number of samples in the fingerprint test set is the number of sample trajectory detection data included in the fingerprint test set before the sample trajectory detection data in the fingerprint test set is added to the second fingerprint library through the second classification model;

The second calculation module is used to divide the added amount by the total sample amount to obtain the trajectory utilization rate, and the classification accuracy and the trajectory utilization rate are used to evaluate the positioning performance of the first classification model.

In the embodiment of the present application, since the first classification model is trained based on the first fingerprint library that has undergone data expansion, and the first fingerprint library is obtained by data expansion of the second fingerprint library based on the fingerprint test set, the amount of data in the first fingerprint library is more reliable, and thus the positioning accuracy of the first classification model is also improved.

It should be noted that: when constructing the fingerprint library provided in the above embodiment, the division of the above functional modules is only used as an example to illustrate. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

The functional units and modules in the above embodiments may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit, and the above integrated units may be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the protection scope of the embodiments of the present application.

The fingerprint database construction device and the fingerprint database construction method provided in the above embodiments belong to the same concept. The specific working process and technical effects of the units and modules in the above embodiments can be found in the method embodiment part and will not be repeated here.

In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from a website site, computer, server or data center by wired (such as: coaxial cable, optical fiber, data subscriber line (Digital Subscriber Line, DSL)) or wireless (such as: infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium may be any available medium that a computer can access, or a data storage device such as a server or data center that includes one or more available media integrations. The available medium may be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a digital versatile disc (DVD)), or a semiconductor medium (such as a solid state disk (SSD)).

The above are optional embodiments provided for the present application and are not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the technical scope disclosed in the present application shall be included in the protection scope of the present application.

Claims (15)

1. An indoor positioning method, characterized in that the method comprises: In response to a positioning operation on a positioning device, acquiring trajectory detection data of the positioning device; Based on the trajectory detection data, determining the floor where the positioning device is located by using a first classification model; The first classification model is obtained by iteratively training the second classification model based on the first fingerprint library in advance, the first fingerprint library is obtained by expanding the data of the second fingerprint library through the second classification model based on the fingerprint test set, the second classification model is obtained by training the initial classification model based on the second fingerprint library, and the second fingerprint library includes multiple sample trajectory detection data, and each of the multiple sample trajectory detection data has a floor label. 2. The method as claimed in claim 1 is characterized in that the first fingerprint library is obtained by traversing the sample trajectory detection data in the fingerprint test set through the second classification model, and during the traversal process, the sample trajectory detection data whose classification result confidence is greater than or equal to the confidence threshold is added to the second fingerprint library. 3. The method according to claim 1 or 2, characterized in that the second fingerprint library is obtained after data expansion processing is performed on the initial fingerprint library based on the initial fingerprint library and the fingerprint test set, and the initial fingerprint library is obtained by randomly selecting a second number of flat layer sample trajectory detection data from the acquired first number of sample trajectory detection data. 4. The method according to claim 3, wherein the fingerprint test set is a set of sample trajectory detection data remaining after randomly selecting from the first number of sample trajectory detection data. 5. A method for constructing a fingerprint library, characterized in that the method comprises: Obtain a second fingerprint library and a fingerprint test set, wherein the second fingerprint library includes a plurality of sample trajectory detection data, and each of the plurality of sample trajectory detection data has a floor label; Iteratively training the initial classification model based on the second fingerprint library to obtain a second classification model; Performing traversal processing on the sample trajectory detection data in the fingerprint test set by using the second classification model; During the traversal process, the sample trajectory detection data whose classification result confidence is greater than or equal to the confidence threshold is added to the second fingerprint library; When the traversal is completed, the second classification model is iteratively trained based on the first fingerprint library to obtain a first classification model. The first fingerprint library is obtained by expanding the second fingerprint library during the traversal process. The first classification model can determine the floor where the positioning device is located based on the trajectory detection data of any positioning device. 6. The method according to claim 5, characterized in that the classification result includes a floor label and a probability value corresponding to the floor label; In the traversal process, adding the sample trajectory detection data whose classification result confidence is greater than or equal to the confidence threshold to the second fingerprint library includes: During the traversal process, according to the probability value in the classification result corresponding to the first sample trajectory detection data, the confidence of the classification result corresponding to the first sample trajectory detection data is determined, and the first sample trajectory detection data is any sample trajectory detection data in the fingerprint test set currently traversed; When the confidence of the classification result corresponding to the first sample trajectory detection data is greater than or equal to the confidence threshold, updating the floor label of the first sample trajectory detection data to the floor label in the classification result corresponding to the first sample trajectory detection data; The first sample trajectory detection data after label update is added to the second fingerprint library. 7. The method according to claim 5, wherein obtaining the second fingerprint library and the fingerprint test set comprises: Acquire a first number of sample trajectory detection data, wherein the first number of sample trajectory detection data at least includes flat layer sample trajectory detection data; Randomly selecting a second number of flat layer sample trajectory detection data from the first number of sample trajectory detection data to obtain an initial fingerprint library; Determine a set consisting of remaining sample trajectory detection data after randomly selecting the first number of sample trajectory detection data as the fingerprint test set; Based on the initial fingerprint library and the fingerprint test set, data expansion processing is performed on the initial fingerprint library to obtain the second fingerprint library. 8. The method according to claim 7, wherein the step of performing data expansion processing on the initial fingerprint library based on the initial fingerprint library and the fingerprint test set comprises: Traversing each sample trajectory detection data in the fingerprint test set in turn; Determine the Euclidean distance and the Jaccard distance between the second sample trajectory detection data and each sample trajectory detection data in the initial fingerprint library, wherein the second sample trajectory detection data is the currently traversed sample trajectory detection data; In the case where there is at least one third sample trajectory detection data in the initial fingerprint library, determining the third sample trajectory detection data having the greatest similarity to the second sample trajectory detection data from the at least one third sample trajectory detection data, wherein the third sample trajectory detection data refers to the sample trajectory detection data having a Euclidean distance with the second sample trajectory detection data less than a first distance threshold and a Jaccard distance with the second sample trajectory detection data less than a second distance threshold; Updating the floor label of the second sample trajectory detection data to the determined floor label of the third sample trajectory detection data; The second sample trajectory detection data after the label is updated is added to the initial fingerprint library. 9. The method according to claim 8, wherein, when there is at least one third sample trajectory detection data in the initial fingerprint library, determining the third sample trajectory detection data having the greatest similarity to the second sample trajectory detection data from the at least one third sample trajectory detection data comprises: In the case where there is at least one third sample trajectory detection data in the initial fingerprint library, obtaining the third sample trajectory detection data having the smallest Euclidean distance with the second sample trajectory detection data from the at least one third sample trajectory detection data; or, In the case that at least one third sample trajectory detection data exists in the initial fingerprint library, the third sample trajectory detection data having the smallest Jaccard distance with the second sample trajectory detection data is obtained from the at least one third sample trajectory detection data. 10. The method according to claim 7, wherein the step of performing data expansion processing on the initial fingerprint library based on the initial fingerprint library and the fingerprint test set comprises: Traversing each sample trajectory detection data in the fingerprint test set in turn; Determine the similarity between the second sample trajectory detection data and each sample trajectory detection data in the initial fingerprint library to obtain multiple similarities, wherein the second sample trajectory detection data is the currently traversed sample trajectory detection data; When the maximum similarity among the multiple similarities is greater than or equal to the similarity threshold, updating the floor label of the second sample trajectory detection data to the floor label of the sample trajectory detection data corresponding to the maximum similarity; The second sample trajectory detection data after the label is updated is added to the initial fingerprint library. 11. The method according to any one of claims 7 to 10, characterized in that before randomly selecting a second number of flat layer sample trajectory detection data from the first number of sample trajectory detection data to obtain an initial fingerprint library, the method further comprises: In a case where the first number of sample trajectory detection data includes sample trajectory detection data whose number of trajectory points is greater than the number threshold, segmenting the sample trajectory detection data whose number of trajectory points is greater than the number threshold to obtain a third number of sample trajectory detection data, wherein the number of trajectory points of each sample trajectory detection data in the third number of sample trajectory detection data is less than or equal to the number threshold; The randomly selecting a second number of flat layer sample trajectory detection data from the first number of sample trajectory detection data to obtain an initial fingerprint library includes: The second number of flat layer sample trajectory detection data is randomly selected from the third number of sample trajectory detection data to obtain the initial fingerprint library. 12. The method according to claim 11, wherein when the first number of sample trajectory detection data includes sample trajectory detection data whose number of trajectory points is greater than a number threshold, segmenting the sample trajectory detection data whose number of trajectory points is greater than the number threshold comprises: In a case where the first number of sample trajectory detection data includes sample trajectory detection data whose number of trajectory points is greater than the number threshold, segmenting the sample trajectory detection data whose number of trajectory points is greater than the number threshold at intervals of a segmentation value to obtain a fourth number of sample trajectory detection data, wherein the segmentation value is a value less than or equal to the number threshold; In the case that the fourth amount of sample trajectory detection data includes sample trajectory detection data whose number of trajectory points is smaller than the division value, the sample trajectory detection data whose number of trajectory points is smaller than the division value is deleted. 13. The method according to claim 5, characterized in that after the sample trajectory detection data in the fingerprint test set is traversed and processed by the second classification model, it also includes: Obtaining the correct amount of sample trajectory detection data in the fingerprint test set classified by the second classification model, and the amount of sample trajectory detection data added to the second fingerprint library after traversing the sample trajectory detection data in the fingerprint test set through the second classification model; The correct amount is divided by the total number of samples in the fingerprint test set to obtain the classification accuracy, where the total number of samples in the fingerprint test set is the number of sample trajectory detection data included in the fingerprint test set before the sample trajectory detection data in the fingerprint test set is added to the second fingerprint library through the second classification model; The added amount is divided by the total number of samples to obtain the track utilization rate. The classification accuracy and the track utilization rate are used to evaluate the positioning performance of the first classification model. 14. An electronic device, characterized in that the electronic device comprises: a processor and a memory, the memory is used to store one or more programs, the one or more programs include instructions, when the processor executes the instructions, the electronic device is used to execute the indoor positioning method as described in any one of claims 1-4, or the electronic device is used to execute the fingerprint library construction method as described in any one of claims 5-13. 15. A computer-readable storage medium for storing one or more programs, wherein the one or more programs are configured to be executed by one or more processors, and the one or more programs include instructions, which enable an electronic device to execute the indoor positioning method described in any one of claims 1-4, or the instructions enable an electronic device to execute the fingerprint library construction method described in any one of claims 5-13.