WO2023005864A1

WO2023005864A1 - Location semantic fingerprint database construction method and related apparatus

Info

Publication number: WO2023005864A1
Application number: PCT/CN2022/107577
Authority: WO
Inventors: 八华峰; 周才发; 唐国斌; 王永亮; 王方松
Original assignee: 华为技术有限公司
Priority date: 2021-07-28
Filing date: 2022-07-25
Publication date: 2023-02-02
Also published as: CN115696195A

Abstract

Disclosed in embodiments of the present application is a location semantic fingerprint database construction method. The method comprises: acquiring a target trajectory, the target trajectory comprising a first set of trajectory points comprising sensor information and/or first GNSS status information, the sensor information being used for indicating that the first set of trajectory points being trajectory points between different horizontal planes, the first GNSS status information being used for indicating that the first set of trajectory points are trajectory points on an entrance and exit between indoors and outdoors, trajectory points in a second set of trajectory points comprising GNSS location information having a confidence level higher than a threshold, and second GNSS status information in a third set of trajectory points being used for indicating that the first set of trajectory points are indoor trajectory points. The location semantic fingerprint database generated on the basis of the information can be used for fingerprint identification and positioning. According to the present application, indoor semantic fingerprint points can be accurately constructed when accurate GNSS positioning information is missing indoors.

Description

A location semantic fingerprint library construction method and related device

This application claims the priority of the Chinese patent application with the application number 202110859719.0 and the title of the invention "a location semantic fingerprint library construction method and related device" submitted to the China Patent Office on July 28, 2021, the entire contents of which are incorporated by reference in this application.

technical field

The present application relates to the field of computers, in particular to a method for constructing a location semantic fingerprint database and related devices.

Background technique

With the popularization, generalization, and dailyization of location-based services, location-based services have become an increasingly indispensable basic demand for people. Most of people's daily activities are indoors, and most of the terminal use and data connection are also indoors, and people's daily activity information is usually related to location and time. Therefore, the demand for location-based services in the mobile Internet era will become more and more bigger and bigger.

The location semantic recognition technology belongs to the technical field of positioning and navigation. Location semantic recognition is mainly to provide users with upper-level service capabilities such as behavior guidance and semantic guidance on the basis of positioning. For example, when the user enters the range of 3-10 meters from the gate of the subway station, the QR code of the subway is popped up for the user in advance. The location semantic recognition technology also provides underlying data support for epidemic prevention and health management, user portraits, personalized recommendations and advertisement push. The popularity of smart phones and the abundance of mobile phone sensors provide the possibility for high-precision recognition of location semantics.

When the existing location semantic fingerprint database is constructed, the comparison and identification is carried out by using the global navigation satellite system (global navigation satellite system, GNSS) information and the preset location point or geofence information. Such identification schemes rely on GNSS signal sources. This method cannot be used for the location semantic fingerprint library in indoor scenes where GNSS location information cannot be collected due to signal occlusion.

Contents of the invention

In a first aspect, the present application provides a method for constructing a location semantic fingerprint library, the method comprising:

Obtaining a target trajectory, the target trajectory includes a plurality of trajectory points, and the plurality of trajectory points include a first set of track points, a second set of track points and a third set of track points connected in sequence, wherein the first set of track points Including sensor information and/or first GNSS state information, the sensor information is used to indicate that the first track point set is a track point between different levels, and the first GNSS state information is used to indicate the first track The point set is a track point on the entrance and exit between indoor and outdoor, the track point in the second track point set includes GNSS position information with a confidence level higher than the threshold, and the third track point set includes the second GNSS state information , the second GNSS state information is used to indicate that the first track point set is an indoor track point;

Wherein, the sensor information can be a barometer, and there is a significant difference in air pressure values between the track points included in the first track point set, for example, according to the direction of the first track point, there is a significant difference in the barometric pressure values of the track points included in the first track point set If the change trend from large to small or from small to large, it can be considered that the sensor information included in the first track point set is used to indicate that the first track point set is a track point between different levels, and then determine the first track The set of points is determined as a set of fingerprint points for the transition region between indoors and outdoors. The first GNSS status information may be GNSS status;

Wherein, the trajectory points (the second trajectory point set) including the GNSS position information whose confidence level is higher than the threshold can be determined as the outdoor trajectory points in the target trajectory, and then the second trajectory point set is determined as the fingerprint point of the outdoor semantics gather;

Among them, the trajectory set connected with the first trajectory point set in the target trajectory and not connected with the second trajectory point set can be used as the indoor trajectory point, and then the third trajectory point set is determined as the indoor semantic fingerprint point set ;

Determining the first set of trajectory points as a set of fingerprint points in a transition zone between indoors and outdoors;

Determining the second set of trajectory points as a set of fingerprint points for outdoor semantics;

Determining the third set of trajectory points as a set of fingerprint points for indoor semantics;

A location semantic fingerprint library is constructed according to the first track point set, the second track point set, and the third track point set.

In the embodiment of the present application, the trajectory point of the transition area between indoor and outdoor can be identified through sensor information and/or GNSS state information, the trajectory point of the outdoor area can be identified through GNSS position information, and the transition area will be extended from the outdoor area to And the GNSS state can indicate the track point of the indoor state as the track point of the indoor area, and when the accurate GNSS positioning information is missing indoors, the fingerprint point of the indoor semantics can also be accurately constructed.

In a possible implementation, the distance between the track points in the second track point set and the indoor geo-fence is greater than a first threshold and smaller than a second threshold.

It should be understood that the first threshold may be a value of 10 meters and its vicinity, such as 8 meters, 9 meters, 11 meters, 12 meters, etc., the first threshold may be related to the area of the indoor geo-fence area, and the larger the area of the geo-fence area The larger the value of the first threshold, the greater the value of the second threshold can be 50 meters and nearby values, such as 48 meters, 49 meters, 51 meters, 52 meters, etc. The second threshold can also be related to the indoor geo-fence area The area of the geo-fence is related, and the larger the area of the geo-fence area, the larger the value of the second threshold.

Among them, the geo-fence in this application can have two definitions, the first is a convex polygonal closed area composed of a series of longitude and latitude point sequences; the second is a circular range with a radius R of a certain center point.

The distance from the latitude and longitude points inside the geofence to the geofence is 0;

The calculation method of calculating a location point outside the geofence to the geofence can be:

The first type: Every two adjacent points in the geofence can form a line segment, calculate the distance from the point to each line segment, and take the minimum value as the distance from the point to the geofence.

The second: directly calculate the distance from the point to the center point minus the radius R as the distance from the point to the geofence.

In a possible implementation, the plurality of track points include adjacent first track points and second track points, the first track point includes first wireless signal information, and the first wireless signal information includes network The first identifier of the device, the first identifier is not globally unique; the second track point includes second wireless signal information, and the second wireless signal information includes the first identifier and the first identifier of the network device Two identifications, the second identification has global uniqueness; the method also includes:

Based on the fact that the first wireless signal information does not include a globally unique identifier for the network device, multiplexing the second identifier included in the second track point to the first track point, so that The first wireless signal information includes the second identifier.

In some scenarios, when the acquisition device collects the track points, the wireless signal information of some track points may be missing. Specifically, the wireless signal information may include the network device identifier (the first identifier and the second identifier) of the adjacent cell, wherein, The first identifier does not have global uniqueness, and the second identifier has global uniqueness. The so-called global uniqueness here means that the second identifier can uniquely indicate the cell where it is located, and there are different cells that correspond to the same first identifier. Example Specifically, the first identifier may be a global cell identifier PCI, and the second identifier may be a pseudo cell identifier CGI. Due to information security considerations, when some base stations send wireless signals to terminals, the wireless signal information of the main cell may include the first identifier and the second identifier, while the wireless signal information of the adjacent cell does not include the second identifier, but only includes the first identifier , in this case, the wireless signal information of the track point only includes the first identifier and does not include the second identifier, and since the first identifier cannot uniquely indicate the network device, the track point is not available.

In the embodiment of the present application, based on the wireless signal information of adjacent track points, the second identifier not included in the track points is supplemented, so that the above track points can be used when the location fingerprint database is constructed, and the utilization rate of data is improved. Correspondingly, the positioning accuracy in subsequent position positioning is also improved.

In a possible implementation, the plurality of track points include a first track point, a second track point, and a third track point that are sequentially adjacent, the first track point includes first wireless signal information, and the first track point A wireless signal information includes a first identifier of a network device, and the first identifier is not globally unique; the third track point includes second wireless signal information, and the second wireless signal information includes all The first identifier and the second identifier, the second identifier is globally unique; based on the fact that the first wireless signal information does not include a globally unique identifier for the network device, the third trajectory The second identifier included in the point is multiplexed to the first track point, so that the first wireless signal information includes the second identifier.

In a possible implementation, the plurality of track points include adjacent first track points and second track points, the first track point includes first wireless signal information, and the first wireless signal information includes network The first identification of the device, the first identification is not globally unique; the first track can be obtained, the first track includes the second track point and the fourth track point, the fourth track point includes the second wireless signal information, the second wireless signal information includes the first identifier and the second identifier of the network device, and the second identifier is globally unique;

Based on the fact that the first wireless signal information does not include a globally unique identifier for the network device, multiplexing the second identifier included in the fourth track point to the first track point, so that The first wireless signal information includes the second identifier.

In a possible implementation, the first identity is a pseudo cell identity PCI, and the second identity is a global identifier CGI.

In a possible implementation, each track point includes wireless signal information, and a similarity between wireless signal information included in different track points is smaller than a threshold.

In a possible implementation, due to signal penetration, the wireless signal information of two track points located indoors and outdoors may be very similar, and these track points will reduce the accuracy of subsequent indoor and outdoor positioning. In a possible implementation, the part of the track points in the target track (or other tracks) that meet the above conditions may be eliminated, so that each of the track points in the target track (or other tracks) includes wireless signal information, and The similarity between wireless signal information included in different track points is smaller than a threshold.

Since the track points whose similarity between wireless signal information is greater than the threshold will cause errors in indoor and outdoor recognition when positioning based on the location fingerprint library, in the embodiment of the present application, the track points whose similarity between wireless signal information is greater than the threshold Points are eliminated from the target trajectory, which improves the accuracy of positioning.

In a possible implementation, the third set of track points includes M target track points, and the M target track points are track points in an indoor preset area, and each target track point includes wireless signal information , and the wireless signal information includes the wireless signal strength of at least one network device; the method further includes:

According to the wireless signal strength of each network device included in the M target trajectory points, determine the signal strength distribution feature of each network device, the signal strength distribution feature includes a mapping relationship between wireless signal strength and probability density, so The location semantic fingerprint library includes the signal strength distribution characteristics of each network device. Optionally, the signal intensity distribution feature is a normal distribution feature.

Specifically, statistical analysis may be performed on the wireless signal feature set in the effective area. It is assumed that the signal strength corresponding to each feature satisfies a Gaussian distribution. For estimating the signal strength distribution parameters of each feature in the network device corresponding to the track points of the second track set, a normal distribution N(μ, σ) is used to replace a feature information set. Thus, the storage overhead and the calculation overhead of the subsequent recognition stage are reduced, and data support is provided for high-precision location semantic recognition.

In a possible implementation, the method further includes: acquiring the first location point information collected by the terminal device, where the first location point information includes target wireless signal information, and the target wireless signal information includes information of M network devices Wireless signal strength; according to the wireless signal strength of each network device in the M network devices, determine the corresponding signal strength distribution feature and the target signal strength interval in the signal strength distribution feature from the location semantic fingerprint library, In each of the wireless signal strength intervals, the probability density corresponding to the signal strength distribution feature is higher than a threshold; based on the wireless signal strength of each of the M network devices being within the corresponding target signal strength interval, determine the The terminal device is located in the indoor preset area.

In a possible implementation, the method further includes: based on that the first location point information includes a first target identifier of the target network device and does not include a second target identifier of the target network device, the first target The identifier does not have global uniqueness, the second target identifier has global uniqueness, and the first target identifier including the target network device and the second target identifier including the target network device are determined from the location semantic fingerprint library the second location point information; multiplexing the second target identifier included in the second location point information to the first location point information, so that the first location point information includes the second target identifier.

In a second aspect, the present application provides a device for constructing a location semantic fingerprint library, the device comprising:

An acquisition module, configured to acquire a target trajectory, the target trajectory includes a plurality of trajectory points, and the plurality of trajectory points include a first set of track points, a second set of track points, and a third set of track points connected in sequence, wherein the The first track point set includes sensor information and/or first GNSS state information, the sensor information is used to indicate that the first track point set is a track point between different levels, and the first GNSS state information is used to indicate The first track point set is track points on the entrance and exit between indoor and outdoor, the track points in the second track point set include GNSS position information with a confidence level higher than a threshold, and the third track point set includes second GNSS state information, the second GNSS state information is used to indicate that the first track point set is an indoor track point;

A semantic determination module, configured to determine the first set of trajectory points as a set of fingerprint points in a transition area between indoors and outdoors;

A fingerprint library construction module, configured to construct a location semantic fingerprint library according to the first track point set, the second track point set, and the third track point set.

An embodiment of the present application provides a location semantic fingerprint database construction device, the device includes: an acquisition module, used to acquire the target trajectory, the target trajectory includes a plurality of trajectory points, the plurality of trajectory points are sequentially connected A track point set, a second track point set, and a third track point set, wherein the first track point set includes sensor information and/or first GNSS state information, and the sensor information is used to indicate the first track point The set is track points between different horizontal planes, the first GNSS state information is used to indicate that the first set of track points is track points on the entrance and exit between indoor and outdoor, and the track in the second set of track points The point includes the GNSS position information whose confidence level is higher than the threshold; the semantic determination module is used to determine the first track point set as the fingerprint point set of the transition zone between indoor and outdoor; the second track point set is determined as A fingerprint point set for outdoor semantics; determining the third track point set as a fingerprint point set for indoor semantics; The third set of trajectory points is used to construct a location semantic fingerprint library. In the embodiment of the present application, the trajectory point of the transition area between indoor and outdoor can be identified through sensor information and/or GNSS state information, the trajectory point of the outdoor area can be identified through GNSS position information, and the transition area will be extended from the outdoor area to As the trajectory points of the indoor area, the indoor semantic fingerprint points can also be accurately constructed when accurate GNSS positioning information is missing indoors.

In a possible implementation, the plurality of track points include adjacent first track points and second track points, the first track point includes first wireless signal information, and the first wireless signal information includes network The first identifier of the device, the first identifier is not globally unique; the second track point includes second wireless signal information, and the second wireless signal information includes the first identifier and the first identifier of the network device Two identifications, the second identification has global uniqueness; the device also includes:

an identifier multiplexing module, configured to multiplex the second identifier included in the second track point into the The first track point, so that the first wireless signal information includes the second identifier.

In a possible implementation, the plurality of track points include a first track point, a second track point, and a third track point that are sequentially adjacent, the first track point includes first wireless signal information, and the first track point A wireless signal information includes a first identifier of a network device, and the first identifier is not globally unique; the third track point includes second wireless signal information, and the second wireless signal information includes all The first identification and the second identification, the second identification has global uniqueness; the identification multiplexing module is also used for:

Based on the fact that the first wireless signal information does not include a globally unique identifier for the network device, multiplexing the second identifier included in the third track point to the first track point, so that The first wireless signal information includes the second identifier.

In a possible implementation, the plurality of track points include adjacent first track points and second track points, the first track point includes first wireless signal information, and the first wireless signal information includes network The first identification of the device, the first identification does not have global uniqueness; the identification multiplexing module is also used for:

Acquire a first track, the first track includes the second track point and a fourth track point, the fourth track point includes second wireless signal information, and the second wireless signal information includes all of the network equipment The first identification and the second identification, the second identification is globally unique;

In a possible implementation, the third set of track points includes M target track points, and the M target track points are track points in an indoor preset area, and each target track point includes wireless signal information , and the wireless signal information includes the wireless signal strength of at least one network device; the fingerprint library construction module is specifically used for:

According to the wireless signal strength of each network device included in the M target trajectory points, determine the signal strength distribution feature of each network device, the signal strength distribution feature includes a mapping relationship between wireless signal strength and probability density, so The location semantic fingerprint library includes the signal strength distribution characteristics of each network device.

In a possible implementation, the signal intensity distribution feature is a normal distribution feature.

In a possible implementation, the acquisition module is also used to:

Acquire first location point information collected by the terminal device, where the first location point information includes target wireless signal information, and the target wireless signal information includes wireless signal strengths of M network devices;

The device also includes:

A positioning module, configured to determine a corresponding signal strength distribution feature and a target signal strength interval in the signal strength distribution feature from the location semantic fingerprint library according to the wireless signal strength of each of the M network devices , each of the wireless signal strength intervals has a probability density corresponding to the signal strength distribution feature higher than a threshold;

Based on the wireless signal strength of each of the M network devices being within a corresponding target signal strength interval, it is determined that the terminal device is in the indoor preset area.

In a possible implementation, the device also includes:

An information completion module, configured to include the first target identifier of the target network device and not include the second target identifier of the target network device based on the first location point information, the first target identifier does not have global uniqueness, The second target identifier has global uniqueness, and the second location point information including the first target identifier of the target network device and the second target identifier of the target network device is determined from the location semantic fingerprint library;

The second target identifier included in the second location point information is multiplexed into the first location point information, so that the first location point information includes the second target identifier.

In a third aspect, the present application provides a positioning method, the method comprising:

According to the wireless signal strength of each of the M network devices, determine the corresponding signal strength distribution feature and the target signal strength interval in the signal strength distribution feature from the location semantic fingerprint database, each of the The probability density corresponding to the signal strength distribution feature in the wireless signal strength interval is higher than a threshold;

In a fourth aspect, the embodiment of the present application provides an apparatus for constructing a location semantic fingerprint library, including: one or more processors and memories; wherein, computer-readable instructions are stored in the memories; the one or more processing The computer reads the computer-readable instructions, so that the computer device implements the above-mentioned first aspect and any optional method.

In the fifth aspect, the embodiment of the present application provides a computer-readable storage medium, which is characterized in that it includes computer-readable instructions, and when the computer-readable instructions are run on a computer device, the computer device is made to execute the above-mentioned first aspect. and any of its optional methods.

In the sixth aspect, the embodiment of the present application provides a computer program product, which is characterized in that it includes computer-readable instructions, and when the computer-readable instructions are run on a computer device, the computer device executes the above-mentioned first aspect and its Either method is optional.

In a seventh aspect, the present application provides a chip system, which includes a processor, configured to support an execution device or a training device to implement the functions involved in the above aspect, for example, send or process the data involved in the above method; or, information. In a possible design, the system-on-a-chip further includes a memory, and the memory is used for storing necessary program instructions and data of the execution device or the training device. The system-on-a-chip may consist of chips, or may include chips and other discrete devices.

Description of drawings

Fig. 1 is a schematic diagram of a construction method of a location semantic fingerprint database according to an embodiment of the present application;

Fig. 2 is a schematic diagram of a construction method of a location semantic fingerprint library according to an embodiment of the present application;

FIG. 3 is a schematic diagram of a construction method of a location semantic fingerprint library according to an embodiment of the present application;

FIG. 4 is a schematic diagram of a construction method of a location semantic fingerprint library according to an embodiment of the present application;

FIG. 5 is a schematic diagram of a construction method of a location semantic fingerprint library according to an embodiment of the present application;

FIG. 6 is a schematic diagram of a construction method of a location semantic fingerprint database according to an embodiment of the present application;

FIG. 7 is a schematic diagram of a construction method of a location semantic fingerprint library according to an embodiment of the present application;

FIG. 8 is a schematic diagram of a construction method of a location semantic fingerprint library according to an embodiment of the present application;

FIG. 9 is a schematic diagram of a construction method of a location semantic fingerprint library according to an embodiment of the present application;

FIG. 10 is a schematic diagram of a construction method of a location semantic fingerprint library according to an embodiment of the present application;

FIG. 11 is a schematic diagram of a construction method of a location semantic fingerprint library according to an embodiment of the present application;

FIG. 12 is a schematic diagram of a construction method of a location semantic fingerprint library according to an embodiment of the present application;

FIG. 13 is a schematic diagram of a construction method of a location semantic fingerprint library according to an embodiment of the present application;

FIG. 14 is a schematic diagram of a construction method of a location semantic fingerprint library according to an embodiment of the present application;

FIG. 15 is a schematic structural diagram of a device for constructing a location-semantic fingerprint library according to an embodiment of the present application;

FIG. 16 is a schematic structural diagram of a server provided by an embodiment of the present application;

FIG. 17 is a schematic structural diagram of a chip provided by an embodiment of the present application.

Detailed ways

Embodiments of the present invention will be described below with reference to the drawings in the embodiments of the present invention. The terms used in the embodiments of the present invention are only used to explain specific examples of the present invention, and are not intended to limit the present invention.

The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned accompanying drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the terms used like this It can be interchanged under appropriate circumstances, and this is only to describe the distinguishing method adopted when describing the object of the same attribute in the embodiments of the application.In addition, the terms "comprising" and "having" and any deformation thereof are intended to be Covers a non-exclusive inclusion such that a process, method, system, product, or apparatus comprising a series of elements is not necessarily limited to those elements, but may include other elements not expressly listed or inherent to the process, method, product, or apparatus .

The key terms and abbreviations involved in the embodiments of the present application are briefly explained below, as follows:

Fingerprint: Use wireless signals (such as terminal base station signals, WIFI signals for wireless LAN communication), ubiquitous geomagnetic signals, or small-range signals emitted by deployed Bluetooth signal tags, etc., to measure and record these signals The unique identification name of the signal point (for example: Medium Access Control (MAC) address), signal strength, and the mapped longitude and latitude location point coordinates, and store these record information in the database as a follow-up real-time positioning matching reference, these Each stored location point and corresponding record information may be called a "fingerprint". In the process of real-time positioning, the real-time location can be obtained through successful "fingerprint" matching, and the positioning function can be realized.

Location characteristics: Nearly stable location points in the main indoor structure, including entry and exit points, doors, elevators, stairs, escalators, corridors, open areas, corners, etc.

Movement characteristics: the typical movement characteristics of the user, which may include movement characteristics such as standing still, walking, running, turning, going up/down stairs, etc.

Map matching: based on the identified location features and actual running trajectory, matching the location points provided by the map and the connection relationship between different areas in the map, and correcting the actual moving route and positioning points to the accurate path and location points.

Received signal strength (received signal strength, RSS): specifically refers to the wideband received power received by the terminal on the channel bandwidth, unit dBm, this value is a relative value, the size is related to the terminal receiving antenna quality, surrounding environment link occlusion, and signal transmission related to the distance between the sources.

Dynamic time warping (dynamic time warping, DTW): Based on the idea of dynamic programming, it solves the problem of template matching with different pronunciation lengths. It is an earlier and more classic algorithm in speech recognition. There are almost no Additional calculations are required. In the application of geomagnetic matching, this algorithm is adopted to solve the problem of pulling up or compressing the signal caused by different user speeds during the real-time positioning process, so as to ensure the correct matching with the original data.

Pedestrian dead reckoning (PDR): A method for estimating the distance and direction of pedestrian movement based on the characteristics of human walking dynamics, including step detection, step estimation, and heading estimation. The estimation is realized by using the built-in sensors of the terminal, such as accelerometer, magnetometer, gyroscope, etc.

K nearest neighbor algorithm (K-NearestNeighbor, KNN): Each sample can be represented by its nearest k neighbors. If most of them belong to a certain category, the sample also belongs to this category and has the characteristics of samples in this category.

Weighted K-Nearest Neighbor (WKNN): Aiming at the differences in the files themselves, a weight factor is added to describe the impact of these differences on the results, so as to promote the effect of classification. The algorithm implementation is still equivalent to the KNN algorithm.

Support vector machine algorithm (support vector machine, SVM): In the field of machine learning, it is a supervised learning model, usually used for pattern recognition, classification and regression analysis.

Point of interest (POI): In a geographic information system, a POI can be a house, a store, a mailbox, a bus stop, etc. Each POI contains four aspects of information, name, category, longitude, latitude.

The application scenarios of indoor positioning can be divided into two categories. One is for consumers, including shopping guides in shopping malls, reverse car search, anti-scattering of family members, self-guided tour guides in museum exhibition halls, location guidance for hospitals, scenic spots, airports, etc. Location query and navigation, location sharing, etc.; the other category is for enterprise customers, including crowd monitoring, user behavior analysis, smart storage, business optimization analysis, advertisement push, emergency rescue, etc.

The existing indoor positioning methods mainly have two modes, one is to rely on actively deployed signal tags, for example: signal tags can include radio frequency identification systems, bluetooth tags, infrared emission tags, etc., and the terminal receives the signals sent by these signal tags , associated with the location corresponding to the signal label, so as to calculate the location of the terminal itself. However, the application of signal tags is limited by high deployment costs, and due to the impact of battery life, high O&M costs that require periodic maintenance and replacement. Another mode is to use ubiquitous wireless signals, for example: wireless signals may include base station signals for terminal communication, wireless fidelity (wireless fidelity, WIFI) signals for wireless local area network communication, and these received signal strengths (received signal strength (RSS) as the "fingerprint" of each location, a large-scale collection, classification, and storage of the fingerprint list and corresponding location information of these locations in advance to form a fingerprint database. In the subsequent positioning, the fingerprint of the unknown location is used to match with the fingerprint database, and the location information corresponding to the most matched fingerprint is output as the positioning result. This mode is widely used because of the ubiquitous WIFI hotspots and no hardware cost.

Whether indoor positioning is based on the active deployment of signal tags or based on existing signal tags, or the mode of offline fingerprint collection and online fingerprint matching positioning, there is inevitably an initial investment. Various application solutions are now exploring how to update or maintain existing signal tags or fingerprint databases to ensure the accuracy of these benchmark information, thereby ensuring the accuracy of real-time positioning.

Scenario 1: An underground parking lot with an area of about 10,000 square meters adopts the industry's low-power Bluetooth positioning solution and deploys about 300 signal tags at intervals of 6 meters. One year later, the signal label near the toilet water room will greatly affect the battery life due to the humid environment, causing the power to run out in advance, then this area will become a positioning blind spot; and the location 10 meters away from the toilet water room, due to Bluetooth Positioning requires at least three valid signal tag data, which also affects the positioning of this area; after two or three years, the number of effective tags is only 80%. As time goes by, the number of effective signal tags will become more and more limited, and a single signal tag will fail. The affected area is within the range of 10 meters * 10 meters. With the failure of multiple signal tags, the positioning effect corresponding to the entire underground parking area will become worse and worse. If the accuracy of the initial Bluetooth positioning is better than 3 meters, it is very important for the signal tag to continue to work effectively. For this reason, it is necessary to periodically replace the old and damaged signal tags and update the fingerprint database synchronously.

For the positioning method of actively deploying signal tags, the existing solution is mainly to update and maintain regularly. The maintenance cycle is based on the battery life time approved by the signal tag, which is generally four years.

Scenario 2: There are a large number of shops and merchants in a shopping mall. Many shops have been relocated within half a year, and the WIFI hotspots have also been removed; or, some merchants are still in the same shopping mall but have changed to different areas. The WIFI hotspot has been moved to a different area; or the store and merchant have not changed, but the WIFI hotspot router is damaged, which causes the previous WIFI hotspot to disappear, and a new WIFI hotspot appears in the corresponding location. The situation can be a WIFI hotspot in the shopping mall. Among them, there were 7 WIFI hotspots in this area corresponding to half a year ago, and 4 remained unchanged after half a year, 1 was withdrawn due to relocation, and two of them were replaced due to the change of location. Added 1 new. The resulting problem is: when the positioning terminal is at position A in the lower left corner, the received WIFI hotspot signal strength is matched with the fingerprint database, and the final real-time positioning result will be at position B, that is, the position at the lower left corner A, mistakenly positioned at the lower center position B. Similarly, at position C, even if the terminal can receive a strong WIFI hotspot signal, because the newly added WIFI hotspot has no corresponding location information and fingerprint list in the fingerprint database, the location result cannot be matched eventually. Moreover, with a time interval of half a year, this kind of WIFI hotspot addition, relocation, and deletion is very common in public areas such as large shopping malls, exhibition halls, and airports, which will inevitably lead to inaccurate positioning results.

Existing technical scheme has three classes. The first type uses GNSS information and preset location points or geographic fence information for comparative identification. Such identification schemes rely on GNSS signal sources. This method cannot be used for positioning problems in indoor scenarios where GNSS signals cannot be collected due to signal occlusion. In addition, the GNSS signal source also has the problem of high power consumption, which affects the user's mobile phone battery life and battery life. The second category uses wireless signal sources such as artificially deployed Beacons, such as some indoor value scenarios, to determine whether to push services based on the received signal strength of the target Beacon. This solution has the problem of high cost and cannot be applied on a large scale. The third type uses GNSS information and wireless signal feature correlation information to perform feature correlation on location semantics, so that only wireless signal features are used in the identification process, achieving the goal of not relying on GNSS and reducing power consumption. However, there are two problems in this type of scheme: First, the wireless features are only correlated for outdoor areas with GNSS. Most of the behaviors take place indoors. For example, indoor scenes such as subways and shopping malls cannot collect effective GNSS information due to satellite signal occlusion, so it is impossible to correlate wireless signal features indoors. Secondly, in the recognition stage, only simple wireless signal feature matching is used, and the coverage of the wireless signal itself is used as the range of the actual location semantics, which cannot achieve high-precision recognition.

Based on this, an embodiment of the present application provides a method for constructing a location semantic fingerprint library.

Referring to FIG. 1, FIG. 1 is a schematic flow diagram of a method for constructing a location semantic fingerprint database provided by an embodiment of the present application. As shown in FIG. 1, a method for constructing a location semantic fingerprint database provided by an embodiment of the present application includes:

101. Acquire a target trajectory, the target trajectory includes a plurality of trajectory points, and the multiple trajectory points include a first set of track points, a second set of track points, and a third set of track points connected in sequence, wherein the first track The point set includes sensor information and/or first GNSS state information, the sensor information is used to indicate that the first track point set is a track point between different levels, and the first GNSS state information is used to indicate that the first track point set is a track point between different levels. A track point set is track points on the entrance and exit between indoor and outdoor, the track points in the second track point set include GNSS position information with a confidence level higher than a threshold, and the third track point set includes the second GNSS State information, the second GNSS state information is used to indicate that the first track point set is an indoor track point.

In the embodiment of the present application, the server can obtain the crowdsourcing data reported by the collection device on the terminal side. It can be understood that the collection device here can be a mobile phone, a tablet personal computer, or a laptop computer. , digital camera, personal digital assistant (PDA for short), navigation device, mobile internet device (mobile internet device, MID) or wearable device (wearable device), etc., without specific limitation. In addition, the server can also obtain manually collected trajectory data. It should be understood that, unlike crowdsourcing data, the server does not need to perform steps such as seed identification and growth described later when processing this part of trajectory data. Among them, the collection device on the terminal side can collect available data such as sensors, network signals and global navigation satellite system (global navigation satellite system, GNSS) of the user terminal anonymously without the user's perception through a certain trigger mechanism ( Also known as fingerprint data). Optionally, the sensor signal can include IMU sensor data (such as accelerometer, gyroscope, magnetometer), positioning sensor data (GNSS positioning information, GNSS status (GNSS status)), and wireless signal data can include WiFi, Bluetooth, base station, etc. data.

Optionally, the collection device may have real-time upload capability, that is, the collection device may upload the collected data to the server in real time. Alternatively, after the collection device completes the data collection, it uploads the collected data to the server in a unified manner. The collected data is used by the server to build a fingerprint database.

In the embodiment of this application, the server on the cloud side can perform preprocessing after receiving the original crowdsourcing data. In the preprocessing process, the crowdsourcing data can be checked for validity, as well as indoor and outdoor recognition (indoor outdoor detection, IOD), cross-floor event recognition (cross floor detection, CFD), pedestrian dead reckoning (Pedestrian dead reckoning, PDR) and other data analysis steps to obtain information including indoor and outdoor information, semantic information such as cross-layer events, and relative coordinate information (for example, x-y-z coordinates) and absolute coordinates (for example, longitude-latitude-height). Because the acquisition of each sensor signal is an asynchronous measurement, it is necessary to align the data information of each dimension to the corresponding relative position. This process can be done by interpolating relative positions. The final preprocessing transforms the sensor and wireless network signal information in the crowdsourcing data source into trajectory information composed of ordered coordinate information with wireless features and semantic features. Through preprocessing, crowdsourcing data is extracted as crowdsourcing trajectories with relative location points as the core, taking crowdsourcing trajectories including target trajectories as an example.

In order to identify whether each trajectory point in the target trajectory is a trajectory point in an outdoor area (also referred to as an invalid area in the embodiment of the present application) or a trajectory point in an indoor area (also referred to as a valid area in the embodiment of the application) , is also the trajectory point of the transition area between outdoor and indoor. In addition to obtaining the preprocessed trajectory information, the server can also obtain indoor geo-fence information, which can indicate the physical boundary information of the indoor area. Further, the server Extraction of valid regions and invalid regions can be performed based on the above information. The valid area may be inside the area enclosed by the geo-fence, and the invalid area may be the peripheral area of the geo-fence. Specifically, the server can identify the seed information of valid areas, invalid areas, and transition areas according to the absolute coordinate information contained in the preprocessed trajectory information and semantic information such as cross-layer, indoor and outdoor states.

In a possible implementation, the multiple track points of the target track may include a first track point set, a second track point set, and a third track point set that are sequentially connected.

In a possible implementation, when the trajectory points included in the target trajectory indicate that there is a cross-layer event (for example, when entering a subway station, it is necessary to go from the ground floor to the lower floor, or from the ground floor to a higher ground floor) or there is When a state switching event occurs from outdoor to indoor or from indoor to outdoor, it can be considered that this part of the trajectory points is located in the transition area between outdoor and indoor, and further, the valid area and the invalid area can be divided based on the transition area.

In a possible implementation, aggregate analysis (such as K-means, DBSCAN and other algorithms) can be performed on all indoor and outdoor switching points or absolute coordinate points at cross-layer events, and then the entry and exit points at the edge of the effective area can be identified. These entry and exit points are the transition areas, and the wireless signal feature set here is used as the information of the transition point feature library. It should be understood that the K-means algorithm here can be extended to: weighted K-nearest neighbor algorithm (Weighted K-NearestNeighbor, WKNN), support vector machine algorithm (support vector machine, SVM) and other typical classification algorithms.

102. Determine the first set of track points as a set of fingerprint points in a transition area between indoors and outdoors.

In the embodiment of the present application, the first track point set may be determined as a set of fingerprint points in the transition area between indoor and outdoor, wherein the first track point set may include sensor information and/or GNSS state information, and the sensor information It is used to indicate that the first track point set is a track point between different horizontal planes, and the first GNSS status information is used to indicate that the first track point set is a track point on an entrance and exit between indoors and outdoors.

Wherein, the sensor information can be a barometer, and there is a significant difference in air pressure values between the track points included in the first track point set, for example, according to the direction of the first track point, there is a significant difference in the barometric pressure values of the track points included in the first track point set If the change trend from large to small or from small to large, it can be considered that the sensor information included in the first track point set is used to indicate that the first track point set is a track point between different levels, and then determine the first track The set of points is determined as a set of fingerprint points for the transition region between indoors and outdoors.

103. Determine the second trajectory point set as an outdoor semantic fingerprint point set.

In the embodiment of the present application, the track points (the second set of track points) including the GNSS position information whose confidence level is higher than the threshold in the target track can be determined as the outdoor track points, and then the second set of track points can be determined as the outdoor track points. A collection of semantic fingerprint points.

Specifically, outdoor feature point information with high-confidence GNSS absolute coordinate information and meeting certain preset area conditions can be extracted as invalid area seed information. The conditions that high-confidence GNSS absolute coordinate information need to meet can refer to the following formula:

GNSS.ACC<T _acc ,T _acc ＝10;

Among them, GNSS.ACC is the positioning error in the GNSS data source. The smaller the value, the higher the positioning accuracy. For certain preset area conditions, you can refer to the following distance calculation formula:

distance(point,polygon)∈[D _min ,D _max ], D _min ＝10m, D _max ＝50m;

Among them, the parameters in the distance calculation formula are the longitude and latitude points and the location semantic valid area fence information polygon composed of a series of ordered longitude and latitude points. The output is the distance (meters) from point to polygon. For example, the data within the range of [10m, 50m] can be taken as the invalid area seed data set, and the distance between the track points in the second track point set and the indoor geo-fence is greater than 10 and less than 50.

104. Determine the third trajectory point set as an indoor semantic fingerprint point set.

In the embodiment of the present application, referring to FIG. 2 , the trajectory set connected to the first trajectory point set in the target trajectory and not connected to the second trajectory point set can be used as the trajectory point in the room, and then the third trajectory point set A collection of fingerprint points identified as indoor semantics.

Specifically, it can be determined that the target trajectory is a trajectory set that passes through the transition region and some trajectory points belong to the invalid region, and some of the trajectory points belong to the valid region, and then the part of the wireless signal feature set in the valid region in the target trajectory is regarded as the seed information of the valid region .

In the embodiment of this application, the server can obtain the target trajectory and other trajectories determined by crowdsourcing data except the target trajectory, and use the target trajectory as the seed data to identify whether the trajectory points in other trajectories are outdoor semantics, indoor semantics, or transition Semantics of the region.

In a possible implementation, assuming that there are differences in the characteristics of wireless sensor signals in different rooms, the trajectory information without GNSS information can be mined, specifically by calculating the similarity between wireless signals (such as the difference between signal strengths) Euclidean distance, etc.), assigning the crowdsourced trajectory to the invalid area and valid area determined above.

For all indoor fingerprint information, it can be compared with the second track point set. For indoor tracks that meet certain similarity conditions and have similarity differences with the invalid area feature set, add them to the second track point set. .

With reference to Fig. 3, for the track information of no indoor and outdoor state or GNSS state, carry out similarity comparison with the second set of track points and the third set of track points, the track points meet the higher similarity with the third set of track points, but with When the similarity of the second track point set is low, it is added to the third track point set. When the track point satisfies a higher similarity degree with the second track point set and a lower similarity degree with the third track point set, it can be added to the second track point set. After several rounds of iterations, the effective area seed bank and the invalid area seed bank converge.

Referring to Figure 4, since the characteristic propagation of wireless signals is omnidirectional and can penetrate obstacles such as buildings, there are the same measurable wireless signals in the effective area and the invalid area, making it impossible to distinguish the two. For example, the base station signal in the wireless signal, when there is a high-power signal in the outdoor area, the transmission of the base station to the indoor effective area or the chain network along the rail transit (multiple radio frequency units share a cell) will have the problem of poor distance resolution. Therefore, a comparative analysis scheme of wireless signals is introduced to identify the confusing features that have no difference between the effective area and the invalid area. Construct the identified indifference feature as an invalid feature library, and do not use it as a positional semantic feature to participate in the subsequent process.

For example, refer to FIG. 5 . As shown in FIG. 5 , A is a valid area feature set, and B is an invalid area feature set. The intersection part A∩B of the two is the confusing feature, A-B is the effective feature set of the valid area, and B-A is the effective feature set of the invalid area.

In a possible implementation, when constructing the location fingerprint library, the third track point set includes M target track points, and the M target track points are track points in a preset indoor area, and each of the The target track point includes wireless signal information, and the wireless signal information includes the wireless signal strength of at least one network device, and the wireless signal strength of each network device included in the M target track points can be determined. The signal strength distribution feature, the signal strength distribution feature includes a mapping relationship between wireless signal strength and probability density, and the location semantic fingerprint database includes the signal strength distribution feature of each network device. Referring to FIG. 6 , optionally, the signal intensity distribution characteristic is a normal distribution characteristic.

In one possible implementation, the statistical analysis here can be realized through two schemes: one is to use all valid region samples for parameter estimation. This scheme is suitable for simple location semantic scenarios such as subway station types. One is to cluster the wireless signals in the effective area (such as DBSCAN, Affinity propagation algorithm, etc.), and perform parameter estimation in units of clusters. This solution is suitable for complex scenarios, such as shopping malls.

The final representation of the table structure of the effective feature library of the location semantic valid area can be shown as follows:

Table 1 The structure of feature library table of location semantic valid area

In some scenarios, when the acquisition device collects the track points, the wireless signal information of some track points may be missing. Specifically, the wireless signal information may include the network device identifier (the first identifier and the second identifier) of the adjacent cell, wherein, The first identifier does not have global uniqueness, and the second identifier has global uniqueness. The so-called global uniqueness here means that the second identifier can uniquely indicate the cell where it is located, and there are different cells that correspond to the same first identifier. Example Specifically, the first identifier may be a global cell identifier CGI, and the second identifier may be a pseudo cell identifier PCI. Due to information security considerations, when some base stations send wireless signals to terminals, the wireless signal information of the main cell may include the first identifier and the second identifier, while the wireless signal information of the adjacent cell does not include the second identifier, but only includes the first identifier , in this case, the wireless signal information of the track point only includes the first identifier and does not include the second identifier, and since the first identifier cannot uniquely indicate the network device, the track point is not available.

In a possible implementation, the plurality of track points may include adjacent first track points and second track points, the first track point includes first wireless signal information, and the first wireless signal information includes A first identifier of a network device, where the first identifier is not globally unique; the second track point includes second wireless signal information, and the second wireless signal information includes the first identifier and The second identifier, the second identifier is globally unique; furthermore, based on the fact that the first wireless signal information does not include a globally unique identifier for the network device, all the information included in the second track point may be The second identifier is multiplexed to the first track point, so that the first wireless signal information includes the second identifier. That is to say, there are track points (first track points) with missing information in the target track, and the missing information in the first track points can be supplemented based on adjacent track points (second track points).

Taking the first identity as the pseudo cell identity PCI and the second identity as the global identifier CGI as an example, in a possible implementation, the completion of the CGI mapping relationship is mainly to obtain the base station signal according to the timing relationship information of the base station signal switch relationship between them, thus obtaining the initial mapping table information. This is the advantage brought by crowdsourcing solutions. Crowdsourcing users will carry out daily activities in the preset area during the unaware crowdsourcing process, and the base station will switch according to communication needs and protocols. As shown in Figure 7 and Figure 8, there is a switching relationship between the base station signal Cell1 and Cell2, then there are two such CGI mapping tables:

CGI1+PCI2->CGI2;

CGI2+PCI1->CGI1;

In the same way, through the switching relationship between Cell2 and Cell3, you can also get:

CGI2+PCI3->CGI3;

CGI3+PCI2->CGI2;

In this embodiment of the present application, a CGI mapping table is constructed through time-series base station Cell signal switching information, and complete CGI mapping relationship information is obtained by using a constraint-based connected graph completion scheme. In this way, the missing neighbor cell information is complemented to obtain high-integrity Cell signal feature information.

In a possible implementation, the plurality of track points include a first track point, a second track point, and a third track point that are sequentially adjacent, the first track point includes first wireless signal information, and the first track point A wireless signal information includes a first identifier of a network device, and the first identifier is not globally unique; the third track point includes second wireless signal information, and the second wireless signal information includes all The first identification and the second identification, the second identification is globally unique; further, based on the fact that the first wireless signal information does not include a globally unique identification for the network device, the third The second identifier included in the track point is multiplexed to the first track point, so that the first wireless signal information includes the second identifier.

In a possible implementation, the plurality of track points include adjacent first track points and second track points, the first track point includes first wireless signal information, and the first wireless signal information includes network The first identification of the device, the first identification does not have global uniqueness; the first track can also be obtained, the first track includes the second track point and the fourth track point, and the fourth track point includes the first track point Two wireless signal information, the second wireless signal information includes the first identifier and the second identifier of the network device, and the second identifier is globally unique; further, it may be based on that the first wireless signal information does not include For the globally unique identifier of the network device, multiplexing the second identifier included in the fourth track point to the first track point, so that the first wireless signal information includes the second Two identification.

In a possible implementation, the missing CGI information in the neighboring cell in the original base station (Cell) information can be recovered by using the mapping table. However, there are problems of incomplete CGI mapping relationship table and asymmetry of recovery results by directly using such a mapping table to complement the missing CGI of adjacent stations. In order to solve this problem, in the process of constructing the CGI mapping relationship table, a constrained Unicom graph scheme is introduced to complete the CGI mapping relationship table. Referring to FIG. 9 , the main idea is that if Cell1, Cell2, and Cell3 are included in a base station scanning record, there is a CGI mapping relationship between Cell1 and Cell2, and there is a CGI mapping relationship between Cell2 and Cell3. Then we think that there is such a CGI mapping relationship between Cell1 and Cell3. And complete the side. In this way, the integrity problem of the CGI mapping relationship table is solved.

The schematic structure of the finally generated CGI mapping relationship table can be as follows:

Table 2 CGI mapping relationship table structure

105. Construct a location semantic fingerprint library according to the first track point set, the second track point set, and the third track point set.

In the embodiment of the present application, the location semantic fingerprint library may include multiple fingerprint points (including wireless signal information, indoor and outdoor semantic track points), and the location semantic fingerprint library may be used to determine indoor and outdoor states and specific locations.

Taking the wireless signal as the base station signal as an example, when using the characteristics of the wireless signal of the base station for location semantic recognition, the CGI mapping relationship table can be used to recover the missing CGI information in the adjacent station to obtain complete feature information. For the implementation process, refer to the description of the above-mentioned construction process, and then use the wireless signal feature information to search the location semantic fingerprint database. Obtain all feature information collections of the location semantic fingerprint library. The retrieval process is to query whether the location semantic feature database has a corresponding primary key (Primary Key) according to CGI and PCI.

Afterwards, feature matching of negative log-likelihood (Negative Log-likelihood, NL) distance can be used for positioning. Referring to FIG. The first location point information includes target wireless signal information, and the target wireless signal information includes wireless signal strengths of M network devices; according to the wireless signal strength of each network device in the M network devices, from the location semantic fingerprint Determining corresponding probability densities in the library; determining that the terminal device is in the indoor preset area based on each of the determined probability densities being higher than a corresponding threshold.

Specifically, use the collected wireless signal feature information and the matched location semantic feature information to perform distance estimation:

Among them, d represents the distance calculation result based on NL,

The mean value of the NL distance results of the main cell,

is the mean value of the NL distance results of neighboring cells, and w is the harmonic parameter, with a value of 0.5. How to calculate the NL distance average value NL _avg is described below.

The J _mod (Modified jaccard) distance can be used for normalization when calculating the NL mean. The calculation formula of J _mod is as follows:

Among them, C is a feature set that satisfies a certain distribution range, and only uses the features in the set to participate in distance calculation to improve robustness. This distribution range condition is described by Z-Score less than a certain threshold T _z-score . Z-Score calculation formula is:

Z _N(μ,σ) (RSSI)＝|RSSI-μ|/σ;

The NL distance calculation formula for each feature f _i is as follows:

Among them, μ _i and σ _i are the mean and standard deviation of feature f _i matched in the location semantic feature library. RSSI _i is the signal strength of the wireless signal feature f _i collected in real time. P _N(μ,σ) (x) is the probability density function of the normal distribution N(μ,σ), which can be implemented in two ways:

Implementation one:

Implementation two:

After obtaining the distance estimation result d, confidence information can be obtained through normalized estimation. In this way, the confidence is used for positional semantic recognition and ranking. The confidence estimation function is as follows:

Conf(d)= _PN(0,bw) (d)/ _PN(0,bw) (0), where bw=10;

Among them, bw is the scaling parameter. Filter the location semantics whose confidence is lower than a certain preset threshold (threshold is 0.01).

An embodiment of the present application provides a method for constructing a location semantic fingerprint library, the method comprising: acquiring a target trajectory, the target trajectory including a plurality of trajectory points, the plurality of trajectory points including a set of first trajectory points connected in sequence, The second track point set and the third track point set, wherein the first track point set includes sensor information and/or first GNSS state information, and the sensor information is used to indicate that the first track point set is between different horizontal planes The first GNSS state information is used to indicate that the first set of track points are track points on the entrance and exit between indoor and outdoor, and the track points in the second set of track points include The GNSS position information of the threshold value, the third track point set includes second GNSS state information, and the second GNSS state information is used to indicate that the first track point set is an indoor track point; the first track point set The set of points is determined as a set of fingerprint points in the transition zone between indoor and outdoor; the second track point set is determined as a set of fingerprint points for outdoor semantics; the third set of track points is determined as a set of fingerprint points for indoor semantics; A location semantic fingerprint library is constructed according to the first track point set, the second track point set, and the third track point set. In the embodiment of the present application, the trajectory point of the transition area between indoor and outdoor can be identified through sensor information and/or GNSS state information, the trajectory point of the outdoor area can be identified through GNSS position information, and the transition area will be extended from the outdoor area to As the trajectory points of the indoor area, the indoor semantic fingerprint points can also be accurately constructed when accurate GNSS positioning information is missing indoors.

Next, the fingerprint library construction process in the embodiment of this application will be described in combination with the interaction between the terminal and the cloud and the data processing process:

Referring to Figure 11, the input data on the cloud side comes from the collection on the device side, and may include crowdsourcing data and geographic information of preset areas (such as building outlines, or given locations and ranges). Crowdsourcing data refers to the anonymized collection of sensor, wireless signal and other information of the user's smart terminal through a certain trigger mechanism on the terminal side without the user's perception. Sensor signals include IMU sensor data (accelerometer, gyroscope, magnetometer) and positioning sensor information (GNSS, GNSS Status), etc.; wireless signal data include WiFi, Bluetooth, base station and other information. The building outline information refers to the closed area information described by the orderly latitude and longitude point sequence, which completely and necessarily describes the coverage of the building.

The preprocessing module on the cloud side can perform data verification on sensor data and complete preliminary analysis, extracting semantic features such as indoor and outdoor features, cross-layer events, and location information such as absolute coordinates and relative coordinates. Since multiple data sources are measured asynchronously by different sensors, coordinate interpolation is performed on the characteristics of different wireless signal sensors to ensure the synchronization of input sources. The final preprocessing transforms the sensor and wireless network signal information in the crowdsourcing data source into trajectory information composed of ordered coordinate information with wireless features and semantic features.

The data mining and extraction module can use the preset building outline information to extract trajectory information, identify transition points and data sets of valid and invalid areas. The valid area is the indoor area inside the building outline; the invalid area is the area formed by a certain range of distance bands outside the building outline. The transition point is a set of location points that must be passed when switching between indoor and outdoor areas, such as the entrance and exit of a shopping mall.

The effective feature extraction module based on comparative analysis can compare and analyze the mined wireless network information in valid areas and invalid areas, and obtain specific effective feature sets.

The positional semantic feature calculation module can perform statistical parameter estimation on the effective feature set in the effective area, and use the statistical feature band to replace all feature sets as the positional semantic feature library for output. Reduce storage overhead and provide the possibility to provide high-precision recognition.

The CGI mapping relationship building block can be aimed at the base station as a wireless signal data source. Due to protocol restrictions, the global unique identifier CGI information cannot be obtained in neighboring cells, which affects the recognition accuracy. This patent utilizes the ordered base station signal information in the track to automatically construct a CGI mapping library of the base station's adjacent area, restore the adjacent area's CGI, and thus obtain complete base station information. Note that this module is only for base station wireless signal data sources, and other data sources such as WiFi do not need this module.

Referring to FIG. 12 , when performing location semantic recognition based on the constructed location fingerprint database, the input data may include wireless signal data collected by the device side and reported to the cloud side. The wireless signals here include WiFi, Bluetooth, base stations, etc.

The CGI recovery module can recover the missing data of the adjacent cell by using the CGI mapping relationship library generated during the construction process for the base station data source.

The RFM retrieval module can use the wireless signal to search the location semantic feature library, and obtain all the location semantic feature information related to the wireless signal feature.

The feature matching module based on NL distance can use the wireless network signal data and the matched related location semantic feature information to perform NL-based distance estimation, and obtain confidence information through normalized mapping for all location semantic result distance information, and sort them in turn. Return ordered position semantic results with confidence to the end-side.

Among them, the device-side implementation solution of location semantics can be shown in Figure 13 .

The input data can include the wireless signal data collected by the device side, the CGI mapping relationship library and the city-level location semantic feature library delivered from the cloud side to the device side. The wireless signals here include WiFi, Bluetooth, base stations, etc. The implementation scheme of the subsequent steps is the same as that on the cloud side.

The difference between the device-side implementation scheme and the cloud-side implementation scheme is that the cloud-side implementation scheme reports wireless signals to the cloud side, and all calculations are performed on the cloud side. The device-side implementation solution is to deliver the CGI library and location semantic feature library to the device side, and all calculations are performed on the device side, reducing network delay and improving real-time recognition.

In addition, the embodiment of the present application also provides a positioning method, the method comprising:

In a possible implementation, based on the first location point information, it may also include the first target identifier of the target network device and not include the second target identifier of the target network device, and the first target identifier does not have global uniqueness, the second target identifier has global uniqueness, and the second location including the first target identifier of the target network device and the second target identifier of the target network device is determined from the location semantic fingerprint database Point information: multiplexing the second target identifier included in the second location point information to the first location point information, so that the first location point information includes the second target identifier.

Referring to Fig. 14, Fig. 14 is a schematic structural diagram of an apparatus for constructing a location semantic fingerprint library provided by an embodiment of the present application. The apparatus 1400 may include:

The acquiring module 1401 is configured to acquire a target track, the target track includes a plurality of track points, and the multiple track points include a first set of track points, a second set of track points and a third set of track points connected in sequence, wherein the The first track point set includes sensor information and/or first GNSS state information, the sensor information is used to indicate that the first track point set is a track point between different levels, and the first GNSS state information is used for Indicate that the first track point set is a track point on an entrance and exit between indoors and outdoors, the track points in the second track point set include GNSS position information with a confidence level higher than a threshold, and the third track point set Including second GNSS state information, the second GNSS state information is used to indicate that the first track point set is an indoor track point;

For a specific description of the acquiring module 1401, reference may be made to the description of step 101 in the foregoing embodiment, and details are not repeated here.

A semantic determination module 1402, configured to determine the first set of track points as a set of fingerprint points in a transition area between indoors and outdoors;

For a specific description of the semantic determination module 1402, reference may be made to the description of

steps

102, 103, and 104 in the above-mentioned embodiment, and details are not repeated here.

The fingerprint library construction module 1403 is configured to construct a location semantic fingerprint library according to the first track point set, the second track point set, and the third track point set.

For the specific description of the fingerprint library construction module 1403, reference may be made to the description of step 105 in the above embodiment, and details are not repeated here.

In a possible implementation, the acquisition module is also used to:

The device also includes:

A positioning module, configured to determine a corresponding probability density from the location semantic fingerprint library according to the wireless signal strength of each of the M network devices;

Based on each of the determined probability densities being higher than a corresponding threshold, it is determined that the terminal device is in the indoor preset area.

An embodiment of the present application provides a location semantic fingerprint database construction device, the device includes: an acquisition module, used to acquire the target trajectory, the target trajectory includes a plurality of trajectory points, the plurality of trajectory points are sequentially connected A track point set, a second track point set, and a third track point set, wherein the first track point set includes sensor information and/or first GNSS state information, and the sensor information is used to indicate the first track point The set is track points between different horizontal planes, the first GNSS state information is used to indicate that the first set of track points is track points on the entrance and exit between indoor and outdoor, and the track in the second set of track points The point includes GNSS position information with a confidence level higher than a threshold, the third track point set includes second GNSS state information, and the second GNSS state information is used to indicate that the first track point set is an indoor track point; semantics A determining module, configured to determine the first set of track points as a set of fingerprint points in a transition area between indoor and outdoor; determine the second set of track points as a set of fingerprint points for outdoor semantics; set the third track The point set is determined as an indoor semantic fingerprint point set; a fingerprint library construction module is configured to construct a location semantic fingerprint library according to the first track point set, the second track point set, and the third track point set. In the embodiment of the present application, the trajectory point of the transition area between indoor and outdoor can be identified through sensor information and/or GNSS state information, the trajectory point of the outdoor area can be identified through GNSS position information, and the transition area will be extended from the outdoor area to The trajectory points of the indoor area are used as the trajectory points of the indoor area. When accurate GNSS positioning information is missing indoors, indoor semantic fingerprint points can also be accurately constructed.

As shown in FIG. 15 , it is a schematic diagram of an embodiment of a terminal in the embodiment of the present application. Taking the terminal as a mobile phone as an example for illustration, FIG. 15 shows a block diagram of a partial structure of the mobile phone related to the terminal provided by the embodiment of the present application. 15, the mobile phone includes: a radio frequency (Radio Frequency, RF) circuit 910, a memory 920, an input unit 930, a display unit 940, a sensor 950, an audio circuit 960, a wireless fidelity (wireless fidelity, WIFI) module 970, a processor 980 , and power supply 990 and other components. Those skilled in the art can understand that the structure of the mobile phone shown in FIG. 15 does not constitute a limitation to the mobile phone, and may include more or less components than shown in the figure, or combine some components, or arrange different components.

The following is a specific introduction to each component of the mobile phone in conjunction with Figure 15:

The RF circuit 910 can be used for sending and receiving information or receiving and sending signals during a call. In particular, after receiving the downlink information from the base station, it is processed by the processor 980; in addition, it sends the designed uplink data to the base station. Generally, the RF circuit 910 includes, but is not limited to, an antenna, at least one amplifier, a transceiver, a coupler, a low noise amplifier (Low Noise Amplifier, LNA), a duplexer, and the like. In addition, RF circuitry 910 may also communicate with networks and other devices via wireless communications. The above wireless communication can use any communication standard or protocol, including but not limited to Global System of Mobile communication (Global System of Mobile communication, GSM), General Packet Radio Service (General Packet Radio Service, GPRS), Code Division Multiple Access (Code Division Multiple Access, CDMA), Wideband Code Division Multiple Access (WCDMA), Long Term Evolution (LTE), email, Short Messaging Service (SMS), etc.

The memory 920 can be used to store software programs and modules, and the processor 980 executes various functional applications and data processing of the mobile phone by running the software programs and modules stored in the memory 920 . The memory 920 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application program required by a function (such as a sound playback function, an image playback function, etc.); Data created by the use of mobile phones (such as audio data, phonebook, etc.), etc. In addition, the memory 920 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid-state storage devices.

The input unit 930 can be used to receive input numbers or character information, and generate key signal input related to user settings and function control of the mobile phone. Specifically, the input unit 930 may include a touch panel 931 and other input devices 932 . The touch panel 931, also referred to as a touch screen, can collect touch operations of the user on or near it (for example, the user uses any suitable object or accessory such as a finger or a stylus on the touch panel 931 or near the touch panel 931). operation), and drive the corresponding connection device according to the preset program. Optionally, the touch panel 931 may include two parts, a touch detection device and a touch controller. Among them, the touch detection device detects the user's touch orientation, and detects the signal brought by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device, converts it into contact coordinates, and sends it to the to the processor 980, and can receive and execute commands sent by the processor 980. In addition, the touch panel 931 can be implemented in various types such as resistive, capacitive, infrared, and surface acoustic wave. In addition to the touch panel 931 , the input unit 930 may also include other input devices 932 . Specifically, other input devices 932 may include but not limited to one or more of a physical keyboard, function keys (such as volume control keys, switch keys, etc.), trackball, mouse, joystick, and the like.

The display unit 940 may be used to display information input by or provided to the user and various menus of the mobile phone. The display unit 940 may include a display panel 941. Optionally, the display panel 941 may be configured in the form of a liquid crystal display (Liquid Crystal Display, LCD), an organic light-emitting diode (Organic Light-Emitting Diode, OLED), or the like. Further, the touch panel 931 may cover the display panel 941, and when the touch panel 931 detects a touch operation on or near it, the touch operation is sent to the processor 980 to determine the type of the touch event, and then the processor 980 according to the touch event The type provides a corresponding visual output on the display panel 941 . Although in FIG. 15 , the touch panel 931 and the display panel 941 are used as two independent components to realize the input and input functions of the mobile phone, in some embodiments, the touch panel 931 and the display panel 941 can be integrated to form a mobile phone. Realize the input and output functions of the mobile phone.

The handset may also include at least one sensor 950, such as a light sensor, motion sensor, and other sensors. Specifically, the light sensor can include an ambient light sensor and a proximity sensor, wherein the ambient light sensor can adjust the brightness of the display panel 941 according to the brightness of the ambient light, and the proximity sensor can turn off the display panel 941 and/or when the mobile phone is moved to the ear. or backlight. As a kind of motion sensor, the accelerometer sensor can detect the magnitude of acceleration in various directions (generally three axes), and can detect the magnitude and direction of gravity when it is stationary, and can be used to identify the application of mobile phone posture (such as horizontal and vertical screen switching, related Games, magnetometer attitude calibration), vibration recognition related functions (such as pedometer, tap), etc.; as for other sensors such as gyroscope, barometer, hygrometer, thermometer, infrared sensor, etc. repeat.

The audio circuit 960, the speaker 961, and the microphone 962 can provide an audio interface between the user and the mobile phone. The audio circuit 960 can transmit the electrical signal converted from the received audio data to the speaker 961, and the speaker 961 converts it into an audio signal for output; After being received, it is converted into audio data, and then the audio data is processed by the output processor 980, and then sent to another mobile phone through the RF circuit 910, or the audio data is output to the memory 920 for further processing.

WIFI belongs to the short-distance wireless transmission technology. Through the WIFI module 970, the mobile phone can help users send and receive emails, browse web pages and access streaming media, etc. It provides users with wireless broadband Internet access. Although Fig. 15 shows the WIFI module 970, it can be understood that it is not an essential component of the mobile phone, and can be completely omitted as required without changing the essence of the invention.

The processor 980 is the control center of the mobile phone. It uses various interfaces and lines to connect various parts of the entire mobile phone. By running or executing software programs and/or modules stored in the memory 920, and calling data stored in the memory 920, execution Various functions and processing data of the mobile phone, so as to monitor the mobile phone as a whole. Optionally, the processor 980 may include one or more processing units; preferably, the processor 980 may integrate an application processor and a modem processor, wherein the application processor mainly processes operating systems, user interfaces, and application programs, etc. , the modem processor mainly handles wireless communications. It can be understood that, the foregoing modem processor may not be integrated into the processor 980 .

The mobile phone also includes a power supply 990 (such as a battery) for supplying power to each component. Preferably, the power supply can be logically connected to the processor 980 through the power management system, so that functions such as charging, discharging, and power consumption management can be realized through the power management system.

Although not shown, the mobile phone may also include a camera, a Bluetooth module, etc., which will not be repeated here.

The steps performed by the terminal in the foregoing method embodiments may be based on the terminal structure shown in FIG. 15 , and will not be repeated here.

The embodiment of the present application also provides a server, please refer to Fig. 16, Fig. 16 is a schematic structural diagram of the server provided in the embodiment of the present application, the server may have relatively large differences due to different configurations or performances, and may include one or More than one central processing unit (central processing units, CPU) 1622 (for example, one or more processors) and memory 1632, one or more storage media 1630 for storing application programs 1642 or data 1644 (for example, one or more mass storage equipment). Wherein, the memory 1632 and the storage medium 1630 may be temporary storage or persistent storage. The program stored in the storage medium 1630 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations on the training device. Furthermore, the central processing unit 1622 may be configured to communicate with the storage medium 1630 , and execute a series of instruction operations in the storage medium 1630 on the server 1600 .

The server 1600 can also include one or more power supplies 1626, one or more wired or wireless network interfaces 1650, one or more input and output interfaces 1658, and/or, one or more operating systems 1641, such as Windows Server™, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, etc.

Specifically, the server may acquire a target track, the target track includes a plurality of track points, and the multiple track points include a first set of track points, a second set of track points, and a third set of track points connected in sequence, wherein the The first track point set includes sensor information and/or first GNSS state information, the sensor information is used to indicate that the first track point set is a track point between different levels, and the first GNSS state information is used to indicate The first track point set is track points on the entrance and exit between indoor and outdoor, the track points in the second track point set include GNSS position information with a confidence level higher than a threshold, and the third track point set includes second GNSS state information, the second GNSS state information is used to indicate that the first track point set is an indoor track point;

In the embodiment of the present application, the trajectory point of the transition area between indoor and outdoor can be identified through sensor information and/or GNSS state information, the trajectory point of the outdoor area can be identified through GNSS position information, and the transition area will be extended from the outdoor area to As the trajectory points of the indoor area, the indoor semantic fingerprint points can also be accurately constructed when accurate GNSS positioning information is missing indoors.

In a possible implementation, the method further includes: based on that the first location point information includes a first target identifier of the target network device and does not include a second target identifier of the target network device, the first target The identifier does not have global uniqueness, the second target identifier has global uniqueness, and the first target identifier including the target network device and the second target identifier including the target network device are determined from the location semantic fingerprint database the second location point information; multiplexing the second target identifier included in the second location point information to the first location point information, so that the first location point information includes the second target identifier.

The embodiment of the present application also provides a computer program product, which, when running on a computer, causes the computer to execute the location semantic fingerprint database construction method described in the above embodiments.

An embodiment of the present application also provides a computer-readable storage medium, the computer-readable storage medium stores a program for performing signal processing, and when it is run on a computer, the computer executes the location as described in the above-mentioned embodiments Construction method of semantic fingerprint library.

The function adjustment device provided in the embodiment of the present application may specifically be a chip. The chip includes: a processing unit and a communication unit. The processing unit may be, for example, a processor, and the communication unit may be, for example, an input/output interface, a pin, or a circuit. The processing unit can execute the computer-executed instructions stored in the storage unit, so that the chip in the execution device executes the image enhancement method described in the above embodiment, or the chip in the training device executes the image enhancement method described in the above embodiment. Optionally, the storage unit is a storage unit in the chip, such as a register, a cache, etc., and the storage unit may also be a storage unit located outside the chip in the wireless access device, such as a read-only memory (read- only memory, ROM) or other types of static storage devices that can store static information and instructions, random access memory (random access memory, RAM), etc.

Specifically, please refer to FIG. 17. FIG. 17 is a schematic structural diagram of a chip provided by the embodiment of the present application. The chip can be represented as a neural network processor NPU170, and the NPU 170 is mounted to the main CPU (Host CPU) as a coprocessor. Above, the tasks are assigned by the Host CPU. The core part of the NPU is the operation circuit 1703, and the operation circuit 1703 is controlled by the controller 1704 to extract matrix data in the memory and perform multiplication operations.

In some implementations, the operation circuit 1703 includes multiple processing units (Process Engine, PE). In some implementations, arithmetic circuit 1703 is a two-dimensional systolic array. The arithmetic circuit 1703 may also be a one-dimensional systolic array or other electronic circuits capable of performing mathematical operations such as multiplication and addition. In some implementations, arithmetic circuit 1703 is a general-purpose matrix processor.

For example, suppose there is an input matrix A, a weight matrix B, and an output matrix C. The operation circuit fetches the data corresponding to the matrix B from the weight memory 1702, and caches it in each PE in the operation circuit. The operation circuit takes the data of matrix A from the input memory 1701 and performs matrix operation with matrix B, and the obtained partial or final results of the matrix are stored in the accumulator 1708 .

The unified memory 1706 is used to store input data and output data. The weight data directly accesses the controller (direct memory access controller, DMAC) 1705 through the storage unit, and the DMAC is transferred to the weight storage 1702. Input data is also transferred to unified memory 1706 by DMAC.

The BIU is the Bus Interface Unit, that is, the bus interface unit 1710, which is used for the interaction between the AXI bus and the DMAC and the instruction fetch buffer (Instruction Fetch Buffer, IFB) 1709.

The bus interface unit 1710 (Bus Interface Unit, BIU for short) is used for the instruction fetch memory 1709 to obtain instructions from the external memory, and is also used for the storage unit access controller 1705 to obtain the original data of the input matrix A or the weight matrix B from the external memory.

The DMAC is mainly used to move the input data in the external memory DDR to the unified memory 1706 , to move the weight data to the weight memory 1702 , or to move the input data to the input memory 1701 .

The vector calculation unit 1707 includes a plurality of calculation processing units, and further processes the output of the calculation circuit, such as vector multiplication, vector addition, exponential operation, logarithmic operation, size comparison, etc., if necessary. It is mainly used for non-convolutional/fully connected layer network calculations in neural networks, such as Batch Normalization (batch normalization), pixel-level summation, and upsampling of feature planes.

In some implementations, the vector computation unit 1707 can store the vector of the processed output to unified memory 1706 . For example, the vector calculation unit 1707 may apply a linear function and/or a nonlinear function to the output of the operation circuit 1703, such as performing linear interpolation on the feature plane extracted by the convolutional layer, and for example, a vector of accumulated values to generate an activation value. In some implementations, the vector calculation unit 1707 generates normalized values, pixel-level summed values, or both. In some implementations, the vector of processed outputs can be used as an activation input to operational circuitry 1703, eg, for use in subsequent layers in a neural network.

An instruction fetch buffer (instruction fetch buffer) 1709 connected to the controller 1704 is used to store instructions used by the controller 1704;

The unified memory 1706, the input memory 1701, the weight memory 1702 and the fetch memory 1709 are all On-Chip memories. External memory is private to the NPU hardware architecture.

Wherein, the processor mentioned in any of the above-mentioned places can be a general-purpose central processing unit, a microprocessor, an ASIC, or one or more programs for controlling the relevant steps of the location semantic fingerprint library construction method described in the above-mentioned embodiments implementation of the integrated circuit.

In addition, it should be noted that the device embodiments described above are only illustrative, and the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physically separated. A unit can be located in one place, or it can be distributed to multiple network units. Part or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, in the drawings of the device embodiments provided in the present application, the connection relationship between the modules indicates that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines.

Through the description of the above embodiments, those skilled in the art can clearly understand that the present application can be implemented by means of software plus necessary general-purpose hardware, and of course it can also be realized by special hardware including application-specific integrated circuits, dedicated CPUs, dedicated memories, Special components, etc. to achieve. In general, all functions completed by computer programs can be easily realized by corresponding hardware, and the specific hardware structure used to realize the same function can also be varied, such as analog circuits, digital circuits or special-purpose circuit etc. However, for this application, software program implementation is a better implementation mode in most cases. Based on this understanding, the essence of the technical solution of this application or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product is stored in a readable storage medium, such as a floppy disk of a computer , U disk, mobile hard disk, ROM, RAM, magnetic disk or optical disk, etc., including several instructions to make a computer device (which can be a personal computer, training device, or network device, etc.) execute the method of each embodiment of the present application .

In the above embodiments, all or part of them may be implemented by software, hardware, firmware or any combination thereof. When implemented using software, it may 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 program instructions are loaded and executed on the computer, the processes or functions according to the embodiments of the present application will be generated in whole or in part. The computer can be a general purpose computer, a special purpose computer, a computer network, or other programmable device. The computer instructions may be stored in or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be passed from a website site, computer, training device, or data center Wired (eg, coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (eg, infrared, wireless, microwave, etc.) transmission to another website site, computer, training device, or data center. The computer-readable storage medium may be any available medium that can be stored by a computer, or a data storage device such as a training device or a data center integrated with one or more available media. The available medium may be a magnetic medium (such as a floppy disk, a hard disk, or a magnetic tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid state disk (Solid State Disk, SSD)), etc.

Claims

A method for constructing a location semantic fingerprint library, characterized in that the method comprises:

Obtaining a target trajectory, the target trajectory includes a plurality of trajectory points, and the plurality of trajectory points include a first set of track points, a second set of track points and a third set of track points connected in sequence, wherein the first set of track points Including sensor information and/or first GNSS state information, the sensor information is used to indicate that the first track point set is a track point between different levels, and the first GNSS state information is used to indicate the first track The point set is a track point on the entrance and exit between indoor and outdoor, the track point in the second track point set includes GNSS position information with a confidence level higher than the threshold, and the third track point set includes the second GNSS state information , the second GNSS state information is used to indicate that the first track point set is an indoor track point;

Determining the first set of trajectory points as a set of fingerprint points in a transition zone between indoors and outdoors;

Determining the second set of trajectory points as a set of fingerprint points for outdoor semantics;

Determining the third set of trajectory points as a set of fingerprint points for indoor semantics;

A location semantic fingerprint library is constructed according to the first track point set, the second track point set, and the third track point set.
The method according to claim 1, wherein the distance between the track points in the second set of track points and the indoor geo-fence is greater than a first threshold and smaller than a second threshold.
The method according to claim 1 or 2, wherein the multiple track points include adjacent first track points and second track points, the first track point includes first wireless signal information, and the The first wireless signal information includes a first identifier of the network device, and the first identifier is not globally unique; the second track point includes second wireless signal information, and the second wireless signal information includes the network device's The first identification and the second identification, the second identification is globally unique; the method also includes:

Based on the fact that the first wireless signal information does not include a globally unique identifier for the network device, multiplexing the second identifier included in the second track point to the first track point, so that The first wireless signal information includes the second identifier.
The method according to claim 1 or 2, wherein the plurality of track points include sequentially adjacent first track points, second track points and third track points, and the first track points include the first Wireless signal information, the first wireless signal information includes a first identifier of a network device, and the first identifier is not globally unique; the third track point includes second wireless signal information, and the second wireless signal information The first identification and the second identification of the network device are included, and the second identification is globally unique; the method further includes:

Based on the fact that the first wireless signal information does not include a globally unique identifier for the network device, multiplexing the second identifier included in the third track point to the first track point, so that The first wireless signal information includes the second identifier.
The method according to claim 1 or 2, wherein the multiple track points include adjacent first track points and second track points, the first track point includes first wireless signal information, and the The first wireless signal information includes a first identifier of the network device, and the first identifier is not globally unique; the method further includes:

Acquire a first track, the first track includes the second track point and a fourth track point, the fourth track point includes second wireless signal information, and the second wireless signal information includes all of the network equipment The first identification and the second identification, the second identification is globally unique;

Based on the fact that the first wireless signal information does not include a globally unique identifier for the network device, multiplexing the second identifier included in the fourth track point to the first track point, so that The first wireless signal information includes the second identifier.
The method according to any one of claims 3 to 5, wherein the first identity is a pseudo cell identity PCI, and the second identity is a global identifier CGI.
The method according to any one of claims 1 to 6, wherein each track point includes wireless signal information, and the similarity between wireless signal information included in different track points is smaller than a threshold.
The method according to any one of claims 1 to 7, wherein the third track point set includes M target track points, and the M target track points are track points in an indoor preset area, each The target trajectory point includes wireless signal information, and the wireless signal information includes wireless signal strength of at least one network device; the method further includes:

According to the wireless signal strength of each network device included in the M target trajectory points, determine the signal strength distribution feature of each network device, the signal strength distribution feature includes a mapping relationship between wireless signal strength and probability density, so The location semantic fingerprint library includes the signal strength distribution characteristics of each network device.
The method according to claim 8, characterized in that, the signal intensity distribution characteristic is a normal distribution characteristic.
The method according to claim 8 or 9, wherein the method further comprises:

Acquire the first location point information collected by the terminal device, the first location point information includes target wireless signal information, and the target wireless signal information includes wireless signal strengths of M network devices; according to each of the M network devices For the wireless signal strength of the network device, determine the corresponding signal strength distribution feature and the target signal strength interval in the signal strength distribution feature from the location semantic fingerprint library, each of the wireless signal strength intervals is within the signal strength distribution The probability density corresponding to the feature is higher than the threshold;

Based on the wireless signal strength of each of the M network devices being within a corresponding target signal strength interval, it is determined that the terminal device is in the indoor preset area.
The method according to claim 10, characterized in that the method further comprises:

Based on the fact that the first location point information includes the first target identifier of the target network device and does not include the second target identifier of the target network device, the first target identifier does not have global uniqueness, and the second target identifier has global uniqueness, determining from the location semantic fingerprint database the second location point information including the first target identifier of the target network device and including the second target identifier of the target network device;

The second target identifier included in the second location point information is multiplexed into the first location point information, so that the first location point information includes the second target identifier.
A location semantic fingerprint library construction device, characterized in that the device comprises:

An acquisition module, configured to acquire a target trajectory, the target trajectory includes a plurality of trajectory points, and the plurality of trajectory points include a first set of track points, a second set of track points, and a third set of track points connected in sequence, wherein the The first track point set includes sensor information and/or first GNSS state information, the sensor information is used to indicate that the first track point set is a track point between different levels, and the first GNSS state information is used to indicate The first track point set is track points on the entrance and exit between indoor and outdoor, the track points in the second track point set include GNSS position information with a confidence level higher than a threshold, and the third track point set includes second GNSS state information, the second GNSS state information is used to indicate that the first track point set is an indoor track point;

A semantic determination module, configured to determine the first set of trajectory points as a set of fingerprint points in a transition area between indoors and outdoors;

Determining the second set of trajectory points as a set of fingerprint points for outdoor semantics;

Determining the third set of trajectory points as a set of fingerprint points for indoor semantics;

A fingerprint library construction module, configured to construct a location semantic fingerprint library according to the first track point set, the second track point set, and the third track point set.
The device according to claim 12, wherein the distance between the track points in the second set of track points and the indoor geo-fence is greater than a first threshold and smaller than a second threshold.
The device according to claim 12 or 13, wherein the multiple track points include adjacent first track points and second track points, the first track point includes first wireless signal information, and the The first wireless signal information includes a first identifier of the network device, and the first identifier is not globally unique; the second track point includes second wireless signal information, and the second wireless signal information includes the network device's The first identification and the second identification, the second identification is globally unique; the device also includes:

an identifier multiplexing module, configured to multiplex the second identifier included in the second track point into the The first track point, so that the first wireless signal information includes the second identifier.
The device according to claim 12 or 13, wherein the plurality of track points include sequentially adjacent first track points, second track points and third track points, and the first track points include the first Wireless signal information, the first wireless signal information includes a first identifier of a network device, and the first identifier is not globally unique; the third track point includes second wireless signal information, and the second wireless signal information Including the first identification and the second identification of the network device, the second identification has global uniqueness; the identification multiplexing module is also used for:

Based on the fact that the first wireless signal information does not include a globally unique identifier for the network device, multiplexing the second identifier included in the third track point to the first track point, so that The first wireless signal information includes the second identifier.
The device according to claim 12 or 13, wherein the multiple track points include adjacent first track points and second track points, the first track point includes first wireless signal information, and the The first wireless signal information includes a first identifier of the network device, and the first identifier is not globally unique; the identifier multiplexing module is also used for:

Acquire a first track, the first track includes the second track point and a fourth track point, the fourth track point includes second wireless signal information, and the second wireless signal information includes all of the network equipment The first identification and the second identification, the second identification is globally unique;

Based on the fact that the first wireless signal information does not include a globally unique identifier for the network device, multiplexing the second identifier included in the fourth track point to the first track point, so that The first wireless signal information includes the second identifier.
The device according to any one of claims 14 to 16, wherein the first identifier is a pseudo cell identifier PCI, and the second identifier is a global identifier CGI.
The device according to any one of claims 12 to 17, wherein each track point includes wireless signal information, and the similarity between wireless signal information included in different track points is smaller than a threshold.
The device according to any one of claims 12 to 18, wherein the third track point set includes M target track points, and the M target track points are track points in an indoor preset area, each The target trajectory point includes wireless signal information, and the wireless signal information includes the wireless signal strength of at least one network device; the fingerprint library construction module is specifically used for:

According to the wireless signal strength of each network device included in the M target trajectory points, determine the signal strength distribution feature of each network device, the signal strength distribution feature includes a mapping relationship between wireless signal strength and probability density, so The location semantic fingerprint library includes the signal strength distribution characteristics of each network device.
The device according to claim 19, wherein the signal strength distribution characteristic is a normal distribution characteristic.
The device according to claim 19 or 20, wherein the acquisition module is also used for:

Acquire first location point information collected by the terminal device, where the first location point information includes target wireless signal information, and the target wireless signal information includes wireless signal strengths of M network devices;

The device also includes:

A positioning module, configured to determine a corresponding signal strength distribution feature and a target signal strength interval in the signal strength distribution feature from the location semantic fingerprint library according to the wireless signal strength of each of the M network devices , each of the wireless signal strength intervals has a probability density corresponding to the signal strength distribution feature higher than a threshold;

Based on the wireless signal strength of each of the M network devices being within a corresponding target signal strength interval, it is determined that the terminal device is in the indoor preset area.
The device according to claim 21, further comprising:

An information completion module, configured to include the first target identifier of the target network device and not include the second target identifier of the target network device based on the first location point information, the first target identifier does not have global uniqueness, The second target identifier has global uniqueness, and the second location point information including the first target identifier of the target network device and the second target identifier of the target network device is determined from the location semantic fingerprint database;

The second target identifier included in the second location point information is multiplexed into the first location point information, so that the first location point information includes the second target identifier.
A location semantic fingerprint library construction device, characterized in that it includes: one or more processors and memory; wherein, computer-readable instructions are stored in the memory;

The one or more processors read the computer readable instructions to cause the computer device to implement the method according to any one of claims 1 to 11.
A computer-readable storage medium, characterized in that it includes computer-readable instructions, and when the computer-readable instructions are run on a computer device, the computer device executes the method described in any one of claims 1 to 11 .
A computer program product, characterized by comprising computer-readable instructions, which, when the computer-readable instructions are run on a computer device, cause the computer device to execute the method according to any one of claims 1 to 11.