Positioning method, apparatus, device, and storage medium

Abstract

Embodiments of the present application provide a positioning method, device, equipment and storage medium, wherein the method can comprise: first, obtaining positioning request information of a target object, the positioning request information comprising historical positioning information of the target object, WIFI fingerprint features scanned by a device sending the positioning request information and base station fingerprint features; second, determining a target geographic grid where the target object is located according to the historical positioning information, the WIFI fingerprint features and the base station fingerprint features; and third, determining target positioning information of the target object based on fingerprint features corresponding to the target geographic grid, the fingerprint features comprising fingerprint features of a target WIFI and / or fingerprint features of a target base station that can be scanned in the target geographic grid. The embodiments are used to solve the problem of inaccurate positioning results in the related art.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of geographic information, and in particular to a positioning method and device, equipment and a storage medium. BACKGROUND

[0002] With the continuous development of information technology and terminal equipment, the application scenarios of positioning technology are becoming more and more extensive.

[0003] At present, positioning is divided into two ways, one way is to position based on the global positioning system signal received by the device, and the other way is to position based on the wireless fidelity (WIFI) or base station information scanned by the device as the positioning basis, through network positioning service. How to improve the accuracy of positioning is a problem that needs to be concerned and continuously optimized by those skilled in the art. SUMMARY

[0004] One or more embodiments of the present application describe a positioning method, device, equipment and storage medium to solve the problem of inaccurate positioning results in the related art.

[0005] In order to solve the above technical problems, the present application is implemented as follows:

[0006] According to a first aspect, a positioning method is provided, which can include:

[0007] Obtaining positioning request information of a target object, the positioning request information including historical positioning information of the target object, WIFI fingerprint features scanned by a device sending the positioning request information, and base station fingerprint features;

[0008] Determining a target geographic grid where the target object is located according to the historical positioning information, the WIFI fingerprint features and the base station fingerprint features;

[0009] Determining target positioning information of the target object based on fingerprint features corresponding to the target geographic grid, the fingerprint features including fingerprint features of target WIFI and / or fingerprint features of target base station that can be scanned in the target geographic grid.

[0010] According to a second aspect, a positioning device is provided, which can include:

[0011] An obtaining module for obtaining positioning request information of a target object, the positioning request information including historical positioning information of the target object, WIFI fingerprint features scanned by a device sending the positioning request information, and base station fingerprint features;

[0012] A determining module for determining a target geographic grid where the target object is located according to the historical positioning information, the WIFI fingerprint features and the base station fingerprint features;

[0013] The processing module is configured to determine target positioning information of the target object based on a fingerprint feature corresponding to a target geographic grid, the fingerprint feature including a fingerprint feature of a target WIFI and / or a fingerprint feature of a target base station that are scannable in the target geographic grid.

[0014] According to a third aspect, a computing device is provided, the device comprising at least one processor and a memory, the memory being configured to store computer program instructions, the processor being configured to execute the program of the memory to control the computing device to implement the positioning method according to the first aspect.

[0015] According to a fourth aspect, a computer readable storage medium is provided, the computer readable storage medium storing a computer program, and the computer program, when executed in a computer, causes the computer to execute the positioning method according to the first aspect.

[0016] In the embodiments of the present application, the WIFI fingerprint feature and the base station fingerprint feature scanned by the device sending the current positioning request information and the historical positioning information are obtained, and then the target geographic grid where the target object is located is determined, and then the target positioning information of the target object is determined according to the target geographic grid information corresponding to the target geographic grid. In this way, when the target object is positioned, not only the WIFI and the base station currently connected by the user are considered, but also the historical positioning information of the user is considered, so that the target user is positioned by using multiple positioning bases, and the positioning accuracy of the target positioning information can be greatly improved. BRIEF DESCRIPTION OF DRAWINGS

[0017] The present application can be better understood with reference to the following description of specific embodiments of the present application in conjunction with the attached drawings, in which like reference numerals refer to like elements, and in which:

[0018] Figure 1 A schematic diagram of a positioning method according to an embodiment is shown;

[0019] Figure 2 A flowchart of a positioning method according to an embodiment is shown;

[0020] Figure 3 A schematic diagram of a geographic grid according to an embodiment is shown;

[0021] Figure 4 Another schematic diagram of a geographic grid according to an embodiment is shown;

[0022] Figure 5 A schematic diagram of a structure implementing a positioning method according to an embodiment is shown;

[0023] Figure 6 A flowchart of a structure implementing a positioning method according to an embodiment is shown;

[0024] Figure 7 A structural diagram of a positioning device according to an embodiment is shown;

[0025] Figure 8 A structural diagram of a computing device according to an embodiment is shown. DETAILED DESCRIPTION

[0026] The features and exemplary embodiments of the various aspects of the present application will be described in detail below with reference to the drawings. The following detailed description is merely provided to explain the present application in detail, and is not intended to limit the present application. The present application can be implemented without some of the specific details provided below. The following description of the embodiments is merely provided to provide a better understanding of the present application through showing examples of the present application.

[0027] It should be noted that the relational terms herein such as first and second and the like are used solely to distinguish one entity or action from another, without necessarily requiring or implying any such actual relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without more limitations, an element preceded by "comprises... a" does not, without more constraints, foreclose the existence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0028] To solve the above technical problems, the embodiments of the present application provide a positioning method, device, equipment and storage medium, as shown below.

[0029] First, a positioning architecture provided by the embodiments of the present application is described.

[0030] In a possible embodiment, as shown in Figure 1 When a user performs network positioning, the user can send positioning request information to a server 11 through an electronic device 10, so that the server 11 can receive the positioning request information of the user, which includes historical positioning information of the user, i.e., the electronic device 10, WIFI fingerprint features and base station fingerprint features scanned by the device, i.e., the electronic device 10, sending the positioning request information.

[0031] The historical positioning information can include at least one of the following information: a time interval between determining the historical positioning information and obtaining the positioning request information, a first distance between the historical positioning information and the candidate geographic grid in a first direction, and a second distance between the historical positioning information and the candidate geographic grid in a second direction. Here, the first direction and the second direction are perpendicular to each other in the same dimensional space.

[0032] It should be noted that the first distance and the second distance represent the relative position of the historical positioning information and the device sending the positioning request information, and the relationship between the two is a more generalized trajectory, which does not change with the change of the area. In other positioning methods using absolute position, if the user leaves the model calculation area corresponding to the absolute position, the user's position cannot be obtained, thereby losing the ability to accurately locate.

[0033] Next, according to the WIFI fingerprint feature, the base station fingerprint feature, and the historical positioning information, a target geographic grid in which the target object is located is determined from a plurality of geographic grids. Here, the plurality of geographic grids can divide the map data according to the geographical position, and each geographic grid has corresponding WIFI fingerprint features and / or base station fingerprint features.

[0034] Then, based on the fingerprint feature corresponding to the target geographic grid, target positioning information of the target object is determined, where the fingerprint feature corresponding to the target geographic grid can include a fingerprint feature of a target WIFI and / or a fingerprint feature of a target base station that can be scanned in the target geographic grid.

[0035] In addition, the positioning method provided by the embodiment of the present application can not only position the user indoors through the WIFI fingerprint feature, but also position the outdoor user through the base station fingerprint feature, and at the same time, the accuracy of the target positioning information of the user can be adjusted according to the historical positioning information.

[0036] Therefore, the method provided by the embodiment of the present application provides multiple positioning bases, i.e., the historical positioning information, the WIFI fingerprint feature, and the base station fingerprint feature, to position the target user, so that the application scope of positioning is expanded from outdoor to indoor, the use scenario is expanded, and the positioning accuracy of the target positioning information can be greatly improved.

[0037] In summary, based on the above-mentioned architecture and application scenario, the embodiment of the present application combines Figure 2 The page display method provided by the embodiment of the present application is further described.

[0038] Figure 2 A flowchart of a positioning method according to one embodiment is shown.

[0039] As Figure 2As shown, the method can include steps 210 to 230:

[0040] First, step 210, obtaining positioning request information of a target object, the positioning request information including historical positioning information of the target object, WIFI fingerprint features scanned by a device sending the positioning request information, and base station fingerprint features.

[0041] Second, step 220, determining a target geographic grid where the target object is located according to the historical positioning information, the WIFI fingerprint feature information, and the base station fingerprint features.

[0042] Then, step 230, determining target positioning information of the target object based on fingerprint features corresponding to the target geographic grid, the fingerprint features including fingerprint features of target WIFI and / or fingerprint features of target base stations that can be scanned in the target geographic grid.

[0043] Thus, the positioning method provided by the embodiments of the present application can position users indoors through WIFI fingerprint features and position outdoor users through base station fingerprint features, and can also adjust the accuracy of target positioning information of users according to historical positioning information. In this way, the application range of positioning is expanded from outdoor to indoor, the use scenarios are expanded, and the positioning accuracy of target positioning information is greatly improved.

[0044] The above steps will be described in detail as follows:

[0045] First, step 210, obtaining positioning request information of a target object, the positioning request information including WIFI fingerprint feature information and base station fingerprint features. Here, historical positioning information of the target object before can also be determined according to historical request information of the target object.

[0046] The historical positioning information can include at least one of the following information:

[0047] The time interval between the historical positioning information and the positioning request information, the first distance between the historical positioning information and the candidate geographic grid in a first direction, the second distance between the historical positioning information and the candidate geographic grid in a second direction, historical WIFI information, and / or historical base station information.

[0048] Here, the first direction and the second direction are perpendicular to each other in the same dimensional space. In establishing the relationship between the historical positioning information and the location of the device currently sending the positioning request information, the embodiment of the application uses the first distance and the second distance to describe the relative position of the historical positioning information and the location of each candidate geographic grid. The relative position relationship does not change with the change of the area, ensuring the stability and accuracy of the positioning method provided by the embodiment of the application. In other positioning methods using absolute position, if the user leaves the model calculation area corresponding to the absolute position, the user's location cannot be obtained, and the accurate positioning ability is lost. Therefore, the positioning method provided by the embodiment of the application has wide application range and high adaptability.

[0049] In addition, the candidate geographic grid can refer to a geographic grid related to the WIFI fingerprint feature and the base station fingerprint feature in the positioning request information, such as Figure 3 As shown in the figure, if the geographic grid related to the WIFI fingerprint feature is geographic grid 1, 2 and 3, and the geographic grid related to the base station fingerprint feature is geographic grid 7, 8 and 9, then the candidate geographic grid can include 1, 2, 3, 7, 8 and 9.

[0050] Of course, in actual application scenarios, the historical positioning information can also include at least one of the following information: a place attribute label related to the historical WIFI fingerprint feature or the historical base station fingerprint feature, an attribute label of whether the historical positioning information conforms to high-precision network positioning, a label of whether the target geographic grid includes a base station location, and a time attribute label of historical positioning.

[0051] For example, the place attribute label can be the attribute of the place where the user is, such as a business district, a residential area, an entertainment place, etc.; the time attribute label can be a working day and a non-working day, and can be more specific, such as morning and evening peak, night, etc.

[0052] Secondly, step 220 is involved, and in a possible embodiment, before determining the target geographic grid where the target object is located according to the positioning request information of the target object, the map data needs to be divided into a plurality of geographic grids according to a preset rule, and the specific manner is as follows:

[0053] The map data is divided according to the geographical position to obtain a plurality of geographic grids;

[0054] According to the WIFI fingerprint feature and the base station fingerprint feature corresponding to each geographic grid in the plurality of geographic grids, the geographic grid information corresponding to each geographic grid is determined; wherein the plurality of geographic grids include the target geographic grid.

[0055] Further, the step of dividing the map data according to the geographical position to obtain a plurality of geographic grids involved in the embodiment of the application can specifically include:

[0056] According to geographical positions, the map data is divided by using a latitude-longitude address algorithm (such as a geohash algorithm) and / or a Mercator projection geographical division algorithm to obtain a plurality of geographical grids.

[0057] It should be noted that the size of the geographical grid is used to determine the accuracy of positioning a target object by a fingerprint feature corresponding to the target geographical grid, that is, the size of the geographical grid determines the granularity of positioning. Generally, the smaller the size of the geographical grid, the higher the accuracy of the positioning result. In addition, in addition to the plurality of equal-sized geographical grids shown in the above Figure 3 Figure 4 The size of the geographical grid is adjustable, that is, the user adjusts the size of the geographical grid according to the actual application scenario.

[0058] Based on this, the embodiment of the present application provides a way to determine a target geographical grid by a deep neural network, as shown in Figure 5 The deep neural network mainly includes two stages, the first stage is an offline calculation layer, and the second stage is an online calculation layer. Here, the networks in the offline calculation layer and the online calculation layer can be used as base learners in a learning to rank (LTR) framework, and the base learners can use machine learning methods such as deep learning networks, recurrent neural networks, and decision trees.

[0059] Among them, the offline calculation layer includes an offline feature compression network, and the online calculation layer includes a first ranking network, a second ranking network, and a third ranking network. The online calculation layer calls the offline compressed WIFI fingerprint features and the offline compressed base station fingerprint features in the offline calculation layer to determine the target geographical grid.

[0060] Therefore, the first stage, that is, the generation of the offline feature fingerprint library by the offline feature compression network in the offline calculation layer, will be described below. The offline feature fingerprint library includes offline compressed WIFI fingerprint features and offline compressed base station fingerprint features. The specific process is as follows:

[0061] Obtain the WIFI fingerprint features and the base station fingerprint features corresponding to each geographical grid in a plurality of geographical grids at different geographical positions. Compress the WIFI fingerprint features and the base station fingerprint features corresponding to each geographical grid by using the offline feature compression network to obtain the offline compressed WIFI fingerprint features and the offline compressed base station fingerprint features corresponding to each geographical grid. Generate an offline feature fingerprint library according to the offline compressed WIFI fingerprint features and the offline compressed base station fingerprint features corresponding to a plurality of geographical grids.

[0062] ​The base station information corresponding to each geographic grid includes a plurality of offline original features, and the offline original features include at least one of the following information:

[0063] The number of collection points corresponding to the base station in the geographic grid, the type of the base station, the number of geographic grids covered by the collection points, a first proportion value of the positioning access amount of the geographic grid in the total positioning access amount related to the base station, a second proportion value of the positioning access amount of the geographic grid in the total number of collection points, a signal strength (Received Signal Strength Indication, RSSI) distribution vector of the base station, a key of the base station, and a base station name.

[0064] For example, the base station type can include a Global System for Mobile Communications (gsm) base station, a Code Division Multiple Access (CDMA) base station, a Long Term Evolution (LTE) base station, or a Wideband Code Division Multiple Access (W-CDMA) base station.

[0065] Similarly, the WIFI fingerprint feature corresponding to each geographic grid also includes a plurality of offline original features, and the offline original features include at least one of the following information:

[0066] The number of collection points corresponding to the WIFI in the geographic grid, the type of the WIFI (such as a mobile WIFI or a fixed WIFI), the number of geographic grids covered by the collection points, a third proportion value of the positioning access amount of the geographic grid in the total positioning access amount related to the WIFI, a fourth proportion value of the positioning access amount of the geographic grid in the total number of collection points, a signal strength distribution vector of the WIFI, and a MAC address (Media Access Control Address) of the WIFI.

[0067] The collection points (including the base station fingerprint feature and the WIFI fingerprint feature) are the base station fingerprint features of the electronic device collected when there is a positioning true value. The positioning true value can be a manually collected value or a positioning value obtained by a high-precision positioning method such as GPS.

[0068] It should be noted that the offline original features of the base station fingerprint feature and the offline original features of the WIFI fingerprint feature are features irrelevant to a single positioning request, and the offline original features do not need to be obtained in the electronic device, but can be calculated offline by the server and stored.

[0069] Thus, the format of the final offline compressed features of the offline original features of WIFI or base station, i.e., the offline compressed WIFI fingerprint feature or the offline compressed base station fingerprint feature, is: default features; the coordinates of the geographic grid 1, the offline compressed features of the geographic grid 1, and the size of the geographic grid 1; the coordinates of the geographic grid 2, the offline compressed features of the geographic grid 2, and the size of the geographic grid 2, and so on.

[0070] In addition, in some embodiments, in combination with different geographic grid division methods, the offline original features of the base station fingerprint features and the offline original features of the WIFI fingerprint features corresponding to each geographic grid are different. For example, the length of the same geographic grid and the area of the geographic grid at different latitudes are different in the geographic division algorithm of the Mercator projection, which causes the offline original features of the base station fingerprint features and the offline original features of the WIFI fingerprint features corresponding to the geographic grid to have a proportional relationship with the size of the geographic grid. Thus, when the offline original features of the base station fingerprint features and the offline original features of the WIFI fingerprint features corresponding to the geographic grid are compressed offline, the proportional relationship between the two needs to be considered.

[0071] Of course, in other embodiments, i.e., in the area where the positioning true value of the base station or WIFI does not cover, the offline original features of the base station fingerprint features and the offline original features of the WIFI fingerprint features can also be obtained by the offline compression layer mentioned above.

[0072] Thus, it can be known that the offline original features are compressed by the offline feature compression network, the feature dimension of the offline compressed WIFI fingerprint feature is smaller than the WIFI fingerprint feature information, and the feature dimension of the offline base station information is smaller than the base station fingerprint feature. In this way, the calculation amount is reduced, and the offline compressed information, i.e., the offline compressed WIFI fingerprint feature and the offline compressed base station fingerprint feature, is convenient to retrieve and calculate.

[0073] Thus, the above-mentioned is a way of generating an offline feature fingerprint library by using the offline feature compression network and a plurality of types of offline original features. Based on this, the present embodiment also provides a way of training the offline feature compression network according to the above-mentioned offline original features, i.e., training the offline feature compression network based on the offline original features until the training condition is met, to obtain the trained offline feature compression network. Here, the time of training the offline feature compression network can be when the offline original features are updated, training the offline feature compression network according to the updated offline original features, or periodically training the offline feature compression network.

[0074] Thus, the above various types of offline original features are fused and refined by the offline feature compression layer, that is, not only the signal strength or channel parameter and the like are input into the offline feature compression network, but also various other features such as the position relationship are input, so that the offline feature fingerprint library can provide data support for the online calculation layer in determining the target geographic grid in various scenarios. In addition, the first proportion value, the second proportion value, the third proportion value, the fourth proportion value, the base station type and the type of wireless fidelity are all features that are not easily affected by environmental changes, which improves the positioning accuracy and robustness while taking into account the calculation and storage.

[0075] Based on the above-mentioned offline feature fingerprint library, the step 220 can specifically include:

[0076] obtaining, from the offline feature fingerprint library, offline compressed WIFI fingerprint features corresponding to the WIFI fingerprint feature information in the positioning request information, and offline compressed base station fingerprint features corresponding to the base station fingerprint feature in the positioning request information;

[0077] inputting the offline compressed WIFI fingerprint features, the offline compressed base station fingerprint features and the historical positioning information into a target ranking model to obtain the target geographic grid of the target object; wherein

[0078] The target ranking model is determined by the WIFI fingerprint features and the base station fingerprint features corresponding to each geographic grid in a plurality of geographic grids at different geographic positions.

[0079] Here, according to different ranking networks included in the target ranking model in the embodiments of the present application, the following two ways of obtaining the target geographic grid where the target object is located are provided, as shown below:

[0080] Method one: the target ranking model includes a first ranking network, and only the first ranking network such as a precision ranking network is used to determine the target geographic grid.

[0081] Among them, the offline compressed WIFI fingerprint features, the offline compressed base station fingerprint features and the historical positioning information are input into the first ranking network to obtain a first score of each first geographic grid in a plurality of first geographic grids, wherein the first geographic grid is related to the position of the positioning request information.

[0082] From the plurality of first scores, a target geographic grid corresponding to a first target score satisfying a first preset condition is selected.

[0083] Here, the first ranking network, i.e., the fine ranking network, can adopt a recurrent neural network to establish a spatio-temporal model for modeling the historical positioning information, and for each historical positioning information, five features are generated: a time interval between the historical positioning information and the positioning request information, a first distance between the historical positioning information and the candidate geographic grid in a first direction, a second distance between the historical positioning information and the candidate geographic grid in a second direction, a place attribute label related to the historical WIFI fingerprint feature or the historical base station fingerprint feature, and an attribute label of whether the historical positioning information meets the high-precision network positioning.

[0084] Therefore, for the historical positioning information, a spatio-temporal model can be established using a recurrent neural network to limit the current position corresponding to the positioning request information to a region, and in combination with the offline compressed WIFI fingerprint feature and the offline compressed base station fingerprint feature, the possible positions are ranked to obtain the most likely position of the target object.

[0085] Since at least five-dimensional features are used in the embodiment of the present application, the problem that the current common spatio-temporal model only uses historical trajectory points to learn the law of crowd movement and the problem that the current positioning is inaccurate only using the current positioning basis can be improved, and the positioning accuracy is effectively improved.

[0086] In the second mode, when the target ranking model includes a second ranking network and a third ranking network, the target geographic grid can be determined through the second ranking network and the third ranking network. Here, the third ranking network is the same as the first ranking network in the above first mode, and both are fine ranking networks, and the second ranking network can be a coarse ranking network, and the specific steps are as follows:

[0087] The offline compressed WIFI fingerprint feature, the offline compressed base station fingerprint feature, and the historical positioning information are input into the second ranking network to obtain a second score of each second geographic grid in a plurality of second geographic grids, wherein the second geographic grid is related to the location of the positioning request information;

[0088] From the plurality of second scores, a plurality of candidate geographic grids corresponding to candidate scores satisfying a second preset condition are selected;

[0089] The candidate geographic grid information of each candidate geographic grid in the plurality of candidate geographic grids, the offline compressed WIFI fingerprint feature, the offline compressed base station fingerprint feature, and the historical positioning information are input into the third ranking network to obtain a third score of each candidate geographic grid;

[0090] From the plurality of third scores, a target geographic grid corresponding to a second target score satisfying a third preset condition is selected.

[0091] Here, in the coarse-sorting stage, historical location information is incorporated by manually extracting features. First, the historical location information is divided into time intervals (e.g., 15 seconds, 1 minute, 10 minutes, 30 minutes, 1 hour, etc.). Then, within each interval, the first distance, second distance, and time interval between the historical location information and each second geographic grid are calculated. These statistical indicators can, to some extent, reflect the impact of historical location information on the current location.

[0092] Next, the offline compressed features and metrics related to historical location information are input into the coarse-ranking network. The output of the coarse-ranking network is the second score of each second geographic raster. The raster is then sorted according to the scores, thus completing the coarse-ranking stage. The coarse screening (or mass screening) is the process of selecting candidate geographic rasters from all geographic rasters based on the second scores. Then, the fine-ranking network selects the target geographic raster from the candidate geographic rasters.

[0093] like Figure 6 As shown, the offline computing layer inputs 15 original Wi-Fi features and 20 original base station features into the offline feature compression layer to obtain compressed offline Wi-Fi fingerprint features and compressed offline base station fingerprint features. Each original Wi-Fi feature includes 20 features, and each original base station feature may include 21 features.

[0094] Next, the offline compressed features and indicators related to historical location information are input into the second ranking network, such as filtering 5 candidate geographic rasters from 10 geographic rasters. Then, the above information is input into the third ranking network to determine the target geographic raster from the 5 candidate geographic rasters.

[0095] Therefore, in this embodiment of the invention, the LTR framework is used for positioning. By combining coarse and fine sorting to filter target geographic grids, the positioning accuracy is improved while the computational load is reduced.

[0096] Additionally, in one possible embodiment, prior to step 220, the following may also be included:

[0097] The location time for obtaining historical target location information, which is the most recent information determined before the location request information was obtained;

[0098] If the time interval between the location information of the historical target and the time interval between the location request information are less than or equal to a preset threshold, an indication information is generated. The indication information is used to indicate the target geographic grid where the target object is located based on the WIFI fingerprint feature information, base station fingerprint feature and historical location information.

[0099] Here, the freshness can be understood as the freshness in the online computing layer, that is, the freshness of the WIFI fingerprint feature and the base station fingerprint feature in the positioning request information is calculated according to the last refresh time of the WIFI and the base station from the current time, the greater the freshness value, the farther the last refresh time from the present, the less fresh, and the worse the reliability. When calculating in the online computing layer, the freshness can be used as a feature to give different positioning bases different weights. When the two positioning bases conflict, the fresher WIFI fingerprint feature or base station fingerprint feature will play a greater role.

[0100] Then, step 230 is involved, in a possible embodiment, based on the fingerprint feature, the position point clustering algorithm, the weighted average centroid algorithm or the Fermat point algorithm is used to determine the target positioning information of the target object in the target geographic grid.

[0101] In summary, the positioning method provided by the embodiment of the application not only considers the WIFI fingerprint feature, but also includes the base station fingerprint feature, so that the application range of positioning can be expanded from indoor to indoor and outdoor. Secondly, the WIFI fingerprint feature, the base station fingerprint feature and the historical positioning information effectively improve the problem that the original information is not rich enough, the use scene is limited and the positioning accuracy is low. Then, by inputting various information of the offline feature compression network, the information can be fused and refined to improve the positioning accuracy and expand the use scene.

[0102] In addition, since the offline original feature is single and easily affected by environmental changes, the embodiment of the application provides a plurality of offline original features, including features that are not easily affected by environmental changes, to improve the positioning accuracy and robustness while taking into account the calculation and storage. Finally, the sorting algorithm framework is used for positioning, the generalization ability of the model is improved through the way of coarse sorting combined with fine sorting, and finally the positioning accuracy is improved.

[0103] Based on the above positioning method, the embodiment of the application provides a positioning device with the same principle. Specifically as follows.

[0104] Firstly, Figure 7 A structural block diagram of a positioning device according to one embodiment is shown.

[0105] As Figure 7 shown, the positioning device 700 can specifically include:

[0106] The acquisition module 701 is configured to acquire the positioning request information of the target object, and the positioning request information includes the historical positioning information of the target object, the WIFI fingerprint feature and the base station fingerprint feature scanned by the device sending the positioning request information;

[0107] The determination module 702 is used to determine the target geographic grid where the target object is located based on WIFI fingerprint feature information, base station fingerprint features and historical positioning information;

[0108] The processing module 703 is used to determine the target location information of the target object based on the fingerprint features corresponding to the target geographic grid. The fingerprint features include the fingerprint features of the target WIFI that can be scanned in the target geographic grid and / or the fingerprint features of the target base station.

[0109] Therefore, by acquiring the Wi-Fi fingerprint and base station fingerprint features scanned by the device currently sending the location request information, as well as historical location information, the target geographic grid where the target object is located is determined. Then, based on the target geographic grid information corresponding to the target geographic grid, the target location information of the target object is determined. In this way, when locating the target object, not only the Wi-Fi and base station currently connected to the user are considered, but also the user's historical location information. By using multiple location criteria to locate the target user, the accuracy of the target location information can be greatly improved.

[0110] Based on this, the positioning device 700 will be described in detail below:

[0111] In one possible embodiment, the determining module 702 may be specifically used to obtain offline compressed WIFI fingerprint features corresponding to WIFI fingerprint features, and offline compressed base station fingerprint features corresponding to base station fingerprint features.

[0112] Offline compressed Wi-Fi fingerprint features, offline compressed base station fingerprint features, and historical location information are input into the target ranking model to obtain the target geographic raster of the target object; among which...

[0113] The target ranking model is determined by the WIFI fingerprint features and base station fingerprint features corresponding to each geographic grid in multiple geographic grids of different geographical locations.

[0114] Furthermore, in one possible embodiment, the target ranking model includes a first ranking network, and the determining module 702 can be specifically used to input offline compressed WIFI fingerprint features, offline compressed base station fingerprint features and historical location information into the first ranking network to obtain a first score for each of the multiple first geographic grids, wherein the first geographic grid is related to the location of the location request information.

[0115] From multiple first scores, select the target geographic raster corresponding to the first target score that meets the first preset condition.

[0116] In another possible embodiment, the target ranking model comprises a second ranking network and a third ranking network, and the determining module 702 can be specifically configured to input the offline compressed WIFI fingerprint feature, the offline compressed base station fingerprint feature and the historical positioning information into the second ranking network to obtain a second score of each second geographic grid in a plurality of second geographic grids, wherein the second geographic grid is related to the location of the positioning request information;

[0117] From the plurality of second scores, a plurality of candidate geographic grids corresponding to candidate scores satisfying a second preset condition are selected;

[0118] The candidate geographic grid information, the offline compressed WIFI fingerprint feature, the offline compressed base station fingerprint feature and the historical positioning information of each candidate geographic grid in the plurality of candidate geographic grids are input into the third ranking network to obtain a third score of each candidate geographic grid;

[0119] From the plurality of third scores, a target geographic grid corresponding to a second target score satisfying a third preset condition is selected.

[0120] In the embodiment of the present application, the historical positioning information comprises at least one of the following information:

[0121] The time interval between the historical positioning information and the time of obtaining the positioning request information, the first distance between the historical positioning information and the candidate geographic grid in a first direction, and the second distance between the historical positioning information and the candidate geographic grid in a second direction are determined.

[0122] The first direction and the second direction are perpendicular to each other in the same dimensional space.

[0123] In addition, in some scenarios, the historical positioning information can further comprise at least one of the following information:

[0124] The historical WIFI fingerprint feature and / or the historical base station fingerprint feature, the place attribute label related to the historical WIFI fingerprint feature or the historical base station fingerprint feature, the attribute label of whether the historical positioning information meets the high-precision network positioning, and the label of whether the target geographic grid includes the base station location.

[0125] In addition, the positioning device 700 in the embodiment of the present application can further comprise a generating module 704, and based on this, the obtaining module 701 can be further configured to obtain the positioning time of the historical target positioning information, the historical target positioning information being information determined for the last time before the positioning request information is obtained.

[0126] The generating module 704 is configured to generate indication information, in a case where a time interval between a positioning time of the historical target positioning information and a time of obtaining the positioning request information is less than or equal to a preset threshold, the indication information being used to indicate that the target object is located in the target geographic grid according to the WIFI fingerprint feature information, the base station fingerprint feature and the historical positioning information.

[0127] The base station information includes at least one of the following information:

[0128] The number of collection points corresponding to the base station in the geographic grid, the type of the base station, the number of geographic grids covered by the collection points, a first proportion value of the positioning access amount of the geographic grid in the total positioning access amount related to the base station, a second proportion value of the positioning access amount of the geographic grid in the total number of collection points, and a signal strength distribution vector of the base station.

[0129] The WIFI information includes at least one of the following information:

[0130] The number of collection points corresponding to the WIFI in the geographic grid, the type of the WIFI, the number of geographic grids covered by the collection points, a third proportion value of the positioning access amount of the geographic grid in the total positioning access amount related to the WIFI, a fourth proportion value of the positioning access amount of the geographic grid in the total number of collection points, and a signal strength distribution vector of the WIFI.

[0131] In addition, a feature dimension of the offline compressed WIFI fingerprint feature is less than the WIFI fingerprint feature information, and a feature dimension of the offline base station information is less than the base station fingerprint feature.

[0132] In yet another possible embodiment, the positioning apparatus 700 in the embodiment of the present application can further include a dividing module 705 configured to divide the map data according to geographic positions to obtain a plurality of geographic grids; based on this, the determining module 702 is further configured to determine geographic grid information corresponding to each geographic grid in the plurality of geographic grids according to WIFI fingerprint features and base station fingerprint features corresponding to each geographic grid in the plurality of geographic grids; and the plurality of geographic grids include the target geographic grid.

[0133] The dividing module 705 can be specifically configured to divide the map data according to geographic positions by using a latitude and longitude address algorithm and / or a Mercator projection geographic division algorithm to obtain the plurality of geographic grids.

[0134] The size of the geographic grid is used to determine the accuracy of positioning the target object by using the fingerprint features; and the size of the geographic grid is in an adjustable state.

[0135] In still another possible embodiment, the processing module 703 can be specifically configured to determine the target positioning information of the target object in the target geographic grid by using a position point clustering algorithm, a weighted average centroid algorithm or a Fermat point algorithm based on the fingerprint features.

[0136] To sum up, the positioning device provided by the embodiment of the application not only considers the WIFI fingerprint features scanned by the device sending the positioning request information, but also includes base station fingerprint features, so that the application range of positioning can be expanded from indoor to indoor and outdoor. Secondly, the problems of insufficient original information, limited use scenarios and positioning accuracy are effectively improved through historical positioning information, WIFI fingerprint features and base station fingerprint features. Then, by inputting various information of the offline feature compression network, the information can be fused and refined to improve the positioning accuracy and expand the use scenarios.

[0137] In addition, since the offline original features are single and are easily affected by environmental changes, the embodiment of the application provides various offline original features, including features that are not easily affected by environmental changes, to improve the positioning accuracy and robustness while taking into account the calculation and storage. Finally, the framework of the sorting algorithm is used for positioning, the generalization ability of the model is improved through the way of coarse sorting combined with fine sorting, and finally the positioning accuracy is improved.

[0138] Figure 8 A structural schematic diagram of a computing device according to an embodiment is shown.

[0139] As shown in Figure 8 , a structural diagram of an exemplary hardware architecture of a computing device capable of implementing the positioning method and the positioning device according to the embodiment of the application is shown.

[0140] The computing device can include a processor 801 and a memory 802 storing computer program instructions.

[0141] Specifically, the processor 801 can include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or can be configured as one or more integrated circuits implementing the embodiments of the application.

[0142] The memory 802 can include mass storage for data or instructions. As an example and not by way of limitation, the memory 802 can include a hard disk drive (HDD), a floppy disk drive, flash memory, an optical disc, a magneto-optical disc, magnetic tape, or a Universal Serial Bus (USB) drive or a combination of two or more of these. The memory 802 can include removable or non-removable (or fixed) media, where appropriate. The memory 802 can be internal or external to the integrated gateway device, where appropriate. In particular embodiments, the memory 802 is nonvolatile, solid-state memory. In particular embodiments, the memory 802 includes read-only memory (ROM). Where appropriate, this ROM can be mask-programmed ROM, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), electrically alterable ROM (EAROM), or flash memory, or a combination of two or more of these. In particular embodiments, the memory 802 includes random-access memory (RAM). Where appropriate, this RAM can include single-data rate RAM, double-data rate RAM, or a combination of two or more of these. Although the memory 802 is shown as a single entity in FIG. 8, the memory 802 can include two or more physical memory units. The memory 802 can interface with the processor 801 and / or the interfaces 805a-805n as shown in FIG. 8. Where appropriate, the memory 802 can be on-board the processor 801. As an example, where the processor 801 is a system on a chip, the memory 802 can be on-chip memory controlled by the processor 801.

[0143] The processor 801 implements a positioning method described above in any of the embodiments by reading and executing computer program instructions stored in the memory 802.

[0144] The transceiver 803 is used to implement the communication between the devices in the embodiments of the present application or with other devices.

[0145] In one example, the device can further include a bus 804. As shown in FIG. 8, the processor 801, the memory 802, and the transceiver 803 are connected by the bus 804 and complete the communication between each other. Figure 8

[0146] The bus 804 includes hardware, software, or both. As an example and not by way of limitation, the bus can include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand (IB) interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or another suitable bus or interconnect, or a combination of two or more of these. Where appropriate, the bus 803 can include one or more buses. Although the present application describes and illustrates a particular bus, the present application contemplates any suitable bus or interconnect.

[0147] The embodiments of the present application also provide a computer readable storage medium corresponding to the positioning method described above.

[0148] ​In a possible implementation, the embodiment of the present application provides a computer readable storage medium, which stores a computer program. When the computer program is executed in a computer, the computer is enabled to execute the steps in the positioning method of the embodiment of the present application.

[0149] It should be noted that the present application is not limited to the specific configurations and processes described in the above embodiments and shown in the drawings. For the convenience and brevity of description, the detailed description of known methods is omitted herein, and the specific working processes of the above-described systems, modules and units can refer to the corresponding processes in the foregoing method embodiments, which will not be described herein.

[0150] It can be clearly understood by those skilled in the art that the method processes of the present application are not limited to the specific steps described and shown, and any person skilled in the art can make various changes, modifications and additions, or equivalent replacements and changes in the order of steps, within the technical scope disclosed by the present application, after understanding the spirit of the present application. These modifications or replacements should be covered within the protection scope of the present application.

Claims

1. A positioning method, comprising: Obtain the location request information of the target object, wherein the location request information includes the historical location information of the target object, the WIFI fingerprint features and base station fingerprint features scanned by the device sending the location request information; Obtain the offline compressed WIFI fingerprint feature corresponding to the WIFI fingerprint feature, and the offline compressed base station fingerprint feature corresponding to the base station fingerprint feature, wherein the feature dimension of the offline compressed WIFI fingerprint feature is smaller than that of the WIFI fingerprint feature information; The historical location information, the offline compressed WIFI fingerprint feature, and the offline compressed base station fingerprint feature are input into the target ranking model to determine the target geographic grid where the target object is located; wherein, the target ranking model is determined by the WIFI fingerprint feature and base station fingerprint feature corresponding to each geographic grid in multiple geographic grids of different geographic locations; Based on the fingerprint features corresponding to the target geographic grid, the target location information of the target object is determined. The fingerprint features include the fingerprint features of the target WIFI that can be scanned in the target geographic grid and / or the fingerprint features of the target base station. The size of the geographic grid is used to determine the accuracy of the target geographic grid information in locating the target object, and the size of the geographic grid is adjustable.

2. The method according to claim 1, wherein, The target ranking model includes a first ranking network; The step of inputting the offline compressed Wi-Fi fingerprint feature, the offline compressed base station fingerprint feature, and the historical location information into the target ranking model to obtain the target geographic raster of the target object includes: The offline compressed WIFI fingerprint feature, the offline compressed base station fingerprint feature, and the historical location information are input into the first sorting network to obtain the first score of each first geographic grid in a plurality of first geographic grids, wherein the first geographic grid is related to the location of the location request information; From multiple first scores, select the target geographic raster corresponding to the first target score that meets the first preset condition.

3. The method according to claim 1, wherein, The target ranking model includes a second ranking network and a third ranking network; The step of inputting the offline compressed Wi-Fi fingerprint feature, the offline compressed base station fingerprint feature, and the historical location information into the target ranking model to obtain the target geographic raster of the target object includes: The offline compressed WIFI fingerprint feature, the offline compressed base station fingerprint feature, and the historical location information are input into the second sorting network to obtain a second score for each of the multiple second geographic grids, wherein the second geographic grid is related to the location of the location request information; From multiple second scores, select multiple candidate geographic rasters corresponding to candidate scores that meet the second preset conditions; The fingerprint features corresponding to each of the multiple candidate geographic gratings, the offline compressed WIFI fingerprint features, the offline compressed base station fingerprint features, and the historical location information are input into the third sorting network to obtain the third score of each candidate geographic grating; From multiple third scores, select the target geographic raster corresponding to the second target score that meets the third preset condition.

4. The method according to claim 3, wherein, The historical location information includes at least one of the following: The time interval between the historical location information and the acquisition of the location request information is determined, the first distance between the historical location information and the candidate geographic grid in a first direction is determined, and the second distance between the historical location information and the candidate geographic grid in a second direction is determined. Wherein, the first direction and the second direction are perpendicular to each other in the same dimensional space.

5. The method according to claim 4, wherein, The historical location information also includes at least one of the following: Historical WIFI fingerprint features and / or historical base station fingerprint features, location attribute tags related to the historical WIFI fingerprint features or the historical base station fingerprint features, attribute tags indicating whether the historical location information conforms to high-precision network positioning, and whether the target geographic grid includes a tag indicating the location of the base station.

6. The method according to any one of claims 1-3, wherein, Before determining the target geographic grid where the target object is located based on the historical location information, the WIFI fingerprint features, and the base station fingerprint features, the method further includes: The location time for obtaining historical target location information, wherein the historical target location information is the most recently determined information before obtaining the location request information; If the time interval between the location time of the historical target location information and the location request information is less than or equal to a preset threshold, an indication message is generated; wherein... The indication information is used to indicate the target geographic grid where the target object is located based on the WIFI fingerprint feature, the base station fingerprint feature, and the historical location information.

7. The method according to claim 1, wherein, The base station fingerprint feature includes at least one of the following: The number of collection points corresponding to base stations in the geographic grid, the type of the base station, the number of geographic grids covered by the collection points, the first proportion of the location access volume of the geographic grid to the total location access volume related to the base station, the second proportion of the location access volume of the geographic grid to the total number of collection points, and the signal strength distribution vector of the base station. The WIFI fingerprint feature includes at least one of the following: The number of collection points corresponding to WIFI in the geographic grid, the type of WIFI, the number of geographic grids covered by the collection point, the third proportion of the location access volume of the geographic grid to the total location access volume related to the WIFI, the fourth proportion of the location access volume of the geographic grid to the total number of collection points, and the signal strength distribution vector of the WIFI.

8. The method according to claim 1, wherein, The feature dimension of the offline compressed WIFI fingerprint feature is smaller than that of the WIFI fingerprint feature; the feature dimension of the offline compressed base station fingerprint feature is smaller than that of the base station fingerprint feature.

9. The method according to claim 1, wherein, Before determining the target geographic raster where the target object is located, the method further includes: The map data is divided according to geographical location to obtain multiple geographic grids; Based on the WIFI fingerprint features and base station fingerprint features corresponding to each of the multiple geographic grids, the geographic grid information corresponding to each geographic grid is determined; wherein, the multiple geographic grids include the target geographic grid.

10. The method according to claim 9, wherein, The map data is divided according to geographical location to obtain multiple geographic grids, including: Based on geographical location, the map data is divided into multiple geographic grids using latitude and longitude address algorithms and / or Mercator projection geographic partitioning algorithms.

11. The method according to claim 1, wherein, The step of determining the target location information of the target object based on the fingerprint features corresponding to the target geographic raster includes: Based on the fingerprint features corresponding to the target geographic grid, the target location information of the target object in the target geographic grid is determined by using location point clustering algorithm, weighted average centroid algorithm or Fermat point algorithm.

12. A positioning device, comprising: The acquisition module is used to acquire the location request information of the target object. The location request information includes historical location information, WIFI fingerprint features and base station fingerprint features scanned by the device that sent the location request information. The determination module is used to obtain the offline compressed WIFI fingerprint feature corresponding to the WIFI fingerprint feature and the offline compressed base station fingerprint feature corresponding to the base station fingerprint feature, wherein the feature dimension of the offline compressed WIFI fingerprint feature is smaller than that of the WIFI fingerprint feature information. The historical location information, the offline compressed WIFI fingerprint feature, and the offline compressed base station fingerprint feature are input into the target ranking model to determine the target geographic grid where the target object is located; wherein, the target ranking model is determined by the WIFI fingerprint feature and base station fingerprint feature corresponding to each geographic grid in multiple geographic grids of different geographic locations; The processing module is used to determine the target location information of the target object based on the fingerprint features corresponding to the target geographic grid. The fingerprint features include the fingerprint features of the target WIFI that can be scanned in the target geographic grid and / or the fingerprint features of the target base station. The size of the geographic grid is used to determine the accuracy of the target geographic grid information in locating the target object. The size of the geographic grid is adjustable.

13. A computer device, wherein, The device includes at least one processor and a memory, the memory being used to store computer program instructions, and the processor being used to execute the program in the memory to control the computer device to implement any one of the positioning methods as claimed in claims 1-11.

14. A computer-readable storage medium having a computer program stored thereon, wherein, If the computer program is executed in a computer, the computer is instructed to perform any of the positioning methods as described in claims 1-11.

Images