An outdoor positioning method and device

Abstract

The application provides an outdoor positioning method and device, electronic equipment and computer program product, relates to the technical field of positioning, and the method comprises the following steps: according to the MME real-time signaling data reported by a terminal to be measured, a minimum fingerprint database subset is matched from a pre-stored offline fingerprint database; the similarity of the MME real-time signaling data and each record data in the minimum fingerprint database subset is calculated; a plurality of target record data with the highest similarity are screened out, and real-time positioning information of the terminal to be measured is determined according to the position information of the plurality of target record data. The embodiment of the application effectively improves the real-time performance and accuracy of outdoor positioning.

Description

Technical Field

[0001] This application relates to the field of positioning technology, specifically to an outdoor positioning method, device, electronic equipment, and computer program product. Background Technology

[0002] With the rapid development of information technology and the widespread adoption of various smart devices, the amount of location information available is increasing. The number of upper-layer applications based on user location is also growing, making high-precision outdoor positioning a key research focus. Traditional positioning methods mainly fall into three categories: The first is base station-based positioning. When a user makes a call, accesses the internet, or changes location, the terminal reports its connection logs to the base station. Therefore, the location of the base station the user is currently connected to reflects, to some extent, the user's ambiguous location information. The second method is triangulation based on base station locations. Typically, a user terminal can receive signals from multiple base stations. Triangulation calculates the distance between the base station and the user using the angles of arrival (Angles of Arrival) of the three base station signals, and then uses geometric knowledge to calculate the user's location. The third method is positioning based on MR data. MR data contains feature values ​​such as signal strength. Based on this, an MR feature fingerprint database can be constructed for user positioning.

[0003] Because base stations have a wide coverage area (approximately 1km in urban areas and 2-5km in suburban areas), the first positioning method based on base station location has a relatively large error. The second method, base station triangulation, requires collecting connection records between the user and three base stations simultaneously. However, in reality, not every user has complete data from all three base stations, and the signal angle of arrival data is inaccurate, resulting in lower accuracy for this method as well. The third method suffers from time delays in MR data and low MR acquisition rates in some areas, thus failing to meet real-time positioning requirements and lacking universality.

[0004] In summary, existing outdoor positioning technologies have low real-time performance and accuracy. Summary of the Invention

[0005] This application provides an outdoor positioning method, device, electronic device, and computer program product to solve the problem of low real-time performance and accuracy of existing outdoor positioning technologies, thereby improving the real-time performance and accuracy of outdoor positioning.

[0006] In a first aspect, embodiments of this application provide an outdoor positioning method, including:

[0007] Based on the real-time signaling data of the MME reported by the terminal under test, the smallest subset of the fingerprint database is matched from the pre-stored offline fingerprint database.

[0008] Calculate the similarity between the MME real-time signaling data and each record in the minimum fingerprint subset;

[0009] The system filters out several target record data with the highest similarity and determines the real-time location information of the terminal under test based on the location information of the several target record data.

[0010] The basic data in the offline fingerprint database is generated by associating the MME historical signaling dataset with the location dataset in the second data source.

[0011] In one embodiment, the step of matching the minimum subset of fingerprints from a pre-stored offline fingerprint database based on the MME real-time signaling data reported by the terminal under test includes:

[0012] Based on the MME real-time signaling data reported by the terminal under test, the offline fingerprint database is pruned at least two levels to generate a minimum subset of the fingerprint database that matches the MME real-time signaling data.

[0013] In one embodiment, calculating the similarity between the MME real-time signaling data and each record in the minimum fingerprint subset includes:

[0014] For each record data in the minimum fingerprint subset, calculate the dimensional distance between the MME real-time signaling data and the record data in at least two feature dimensions;

[0015] The similarity between the MME real-time signaling data and the recorded data is determined based on the dimensional distance on at least two feature dimensions.

[0016] In one embodiment, the step of filtering out a number of target record data with the highest similarity and determining the real-time location information of the terminal under test based on the location information of the number of target record data includes:

[0017] Filter out the target record data with the highest similarity and obtain the location information corresponding to the target record data;

[0018] Based on the similarity between each of the target record data and the MME real-time signaling data, a weighted average of the several location information data is performed to determine the real-time location information of the terminal under test.

[0019] In one embodiment, the basic data in the offline fingerprint database is generated by associating the MME historical signaling dataset with the location dataset in the second data source, including:

[0020] Rasterize the location dataset in the second data source;

[0021] The location dataset and the MME historical signaling dataset are matched according to the user ID and time window, and the matching MME historical signaling data records are associated with the corresponding location data records.

[0022] The location raster of the location data record that appears most frequently is used as the precise location raster of the corresponding MME historical signaling data record;

[0023] MME historical signaling data records where the distance from the center point of the precise location grid to the MME base station exceeds a preset threshold are deleted, and the retained MME historical signaling data records are used as the basic data in the offline fingerprint database.

[0024] In one embodiment, the recorded data in the offline fingerprint database is obtained by feature derivation based on the basic data according to a preset feature dimension;

[0025] The preset feature dimensions include one or more of the following: weekday identifier, week identifier, hourly granularity identifier, first time window identifier, and historical motion trajectory identifier.

[0026] In one embodiment, the pruning process on the offline fingerprint database at least two levels includes at least two of the following:

[0027] The offline fingerprint database records that do not match the base station ID of the MME real-time signaling data are filtered out;

[0028] The offline fingerprint database records that do not match the user IDs in the MME real-time signaling data are filtered out.

[0029] The record data in the offline fingerprint database that does not match the weekday label of the MME real-time signaling data is filtered out;

[0030] The offline fingerprint database records that do not match the hourly granularity labels of the MME real-time signaling data are filtered out.

[0031] Records in the offline fingerprint database that do not match the time window labels of the MME real-time signaling data are filtered out.

[0032] Secondly, embodiments of this application provide an outdoor positioning device, comprising:

[0033] The data filtering module is used to match the smallest subset of the fingerprint database from the pre-stored offline fingerprint database based on the real-time signaling data of the MME reported by the terminal under test.

[0034] The data calculation module is used to calculate the similarity between the MME real-time signaling data and each record data in the minimum fingerprint subset;

[0035] The positioning determination module is used to filter out several target record data with the highest similarity, and determine the real-time positioning information of the terminal under test based on the location information of the several target record data.

[0036] The basic data in the offline fingerprint database is generated by associating the MME historical signaling dataset with the location dataset in the second data source.

[0037] Thirdly, embodiments of this application provide an electronic device, including a processor and a memory storing a computer program, wherein the processor executes the program to implement the outdoor positioning method described in the first aspect.

[0038] Fourthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the outdoor positioning method described in the first aspect.

[0039] The outdoor positioning method, apparatus, electronic device, and computer program product provided in this application have the following advantages: the method effectively improves the coverage and precision of the offline fingerprint database by associating and matching MME signaling data with location data from a second data source; it also improves computational efficiency by filtering the offline fingerprint database before performing positioning calculations to reduce the amount of data computation; and thus, the method effectively improves the real-time performance and accuracy of outdoor positioning. Attached Figure Description

[0040] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0041] Figure 1 This is a flowchart illustrating the outdoor positioning method provided in an embodiment of this application;

[0042] Figure 2 This is a schematic diagram of the overall process of fingerprint database positioning provided in the embodiments of this application;

[0043] Figure 3 This is a flowchart illustrating the real-time positioning process provided in an embodiment of this application;

[0044] Figure 4 This is a schematic diagram of the grid position nine-square grid expansion provided in the embodiments of this application;

[0045] Figure 5 This is a schematic diagram of the structure of the outdoor positioning device provided in the embodiments of this application;

[0046] Figure 6 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0047] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0048] Please refer to Figure 1 This application provides an outdoor positioning method, which may include the following steps:

[0049] S1. Based on the MME real-time signaling data reported by the terminal under test, match the smallest subset of the fingerprint database from the pre-stored offline fingerprint database.

[0050] S2. Calculate the similarity between the MME real-time signaling data and each record data in the minimum fingerprint subset;

[0051] S3. Select several target record data with the highest similarity, and determine the real-time location information of the terminal under test based on the location information of the several target record data.

[0052] The basic data in the offline fingerprint database is generated by associating the MME historical signaling dataset with the location dataset in the second data source.

[0053] In this embodiment, the historical signaling dataset of the MME is first associated with the location dataset in the second data source. The low-precision MME signaling data is then corrected using the location dataset from the second data source, serving as the foundation data for the offline fingerprint database. The second data source can be user terminal location data collected from sources such as the MDT data source and the Aladdin data source. Then, the real-time MME signaling data of the terminal under test is matched with the recorded data in the offline fingerprint database according to preset rules. Mismatched data is filtered out, and the remaining recorded data sets are combined into a minimum fingerprint database subset to reduce subsequent computation. The similarity between the real-time MME signaling data and each recorded data in the minimum fingerprint database subset is calculated, sorted from high to low similarity, and one or more target recorded data sets with the highest similarity are selected. Based on the location information of these target recorded data sets, the real-time location information (which can be in latitude and longitude coordinates) of the terminal under test can be determined.

[0054] The outdoor positioning method provided in this application improves the coverage and precision of the offline fingerprint database by associating and matching MME signaling data with location data from a second data source; it also improves computational efficiency by filtering the offline fingerprint database before performing positioning calculations to reduce the amount of data computation; thus, the method effectively improves the real-time performance and accuracy of outdoor positioning.

[0055] In one embodiment, step S1 may include:

[0056] Based on the MME real-time signaling data reported by the terminal under test, the offline fingerprint database is pruned at least two levels to generate a minimum subset of the fingerprint database that matches the MME real-time signaling data.

[0057] Furthermore, the pruning process on the offline fingerprint database, which involves at least two levels, includes at least two of the following:

[0058] The offline fingerprint database records that do not match the base station ID of the MME real-time signaling data are filtered out;

[0059] The offline fingerprint database records that do not match the user IDs in the MME real-time signaling data are filtered out.

[0060] The record data in the offline fingerprint database that does not match the weekday label of the MME real-time signaling data is filtered out;

[0061] The offline fingerprint database records that do not match the hourly granularity labels of the MME real-time signaling data are filtered out.

[0062] Records in the offline fingerprint database that do not match the time window labels of the MME real-time signaling data are filtered out.

[0063] In this embodiment, the offline fingerprint database can be pruned at at least two levels. First, the recorded data in the offline fingerprint database is filtered at the first level (e.g., base station ID), then at the second level (e.g., weekday tag), and so on, to obtain the smallest subset of the pruned fingerprint database. It is understood that the pruning process can be set to two, three, ... N levels. The pruning process at each level can be any of the five items mentioned above, or other methods set according to actual application needs, and the order of pruning at each level can be adjusted as needed. In the above embodiment, the meaning of hourly granularity tag mismatch can be: the hourly granularity tag of the data record in the offline fingerprint database is different from the hourly granularity tag of the MME real-time signaling data, or the hourly granularity tag of the data record in the offline fingerprint database differs from the hourly granularity tag of the MME real-time signaling data by a preset number of hours. In the above embodiment, the time window tag can be set according to actual needs, such as N minutes as a time window.

[0064] The outdoor positioning method provided in this application uses a multi-level filtering mechanism to perform preliminary pruning on the offline fingerprint database, generating a minimum subset of fingerprints that meets the conditions. As a result, subsequent similarity calculations only require traversing and querying within the minimum subset of fingerprints, reducing the amount of computation and effectively improving computational efficiency, thereby further enhancing the real-time performance of outdoor positioning.

[0065] In one embodiment, calculating the similarity between the MME real-time signaling data and each record in the minimum fingerprint subset includes:

[0066] For each record data in the minimum fingerprint subset, calculate the dimensional distance between the MME real-time signaling data and the record data in at least two feature dimensions;

[0067] The similarity between the MME real-time signaling data and the recorded data is determined based on the dimensional distance on at least two feature dimensions.

[0068] In this embodiment of the application, when calculating similarity, the dimensional distance between the MME real-time signaling data and the recorded data can be calculated first on two or more feature dimensions, and then the final distance between the MME real-time signaling data and the recorded data can be further determined based on the two or more dimensional distances (the distances of multiple dimensions can be directly superimposed as the final distance, or a weighted sum can be performed according to the importance of each dimension), thereby determining the similarity between the MME real-time signaling data and the recorded data (the smaller the distance value, the higher the similarity).

[0069] The outdoor positioning method provided in this application improves the universality of similarity calculation by calculating the similarity between MME real-time signaling data and recorded data from multiple dimensions, thereby further improving the accuracy of outdoor positioning.

[0070] In one embodiment, the step of filtering out a number of target record data with the highest similarity and determining the real-time location information of the terminal under test based on the location information of the number of target record data includes:

[0071] Filter out the target record data with the highest similarity and obtain the location information corresponding to the target record data;

[0072] Based on the similarity between each of the target record data and the MME real-time signaling data, a weighted average of the several location information data is performed to determine the real-time location information of the terminal under test.

[0073] In this embodiment, when obtaining multiple location information pieces with the highest similarity, corresponding weights can be set according to the magnitude of each similarity (the higher the similarity, the greater the weight), to perform a weighted average of the multiple location information pieces. This enables the acquisition of more accurate location information, thereby further improving the accuracy of outdoor positioning.

[0074] In one embodiment, the basic data in the offline fingerprint database is generated by associating the MME historical signaling dataset with the location dataset in the second data source, including:

[0075] Rasterize the location dataset in the second data source;

[0076] The location dataset and the MME historical signaling dataset are matched according to the user ID and time window, and the matching MME historical signaling data records are associated with the corresponding location data records.

[0077] The location raster of the location data record that appears most frequently is used as the precise location raster of the corresponding MME historical signaling data record;

[0078] MME historical signaling data records where the distance from the center point of the precise location grid to the MME base station exceeds a preset threshold are deleted, and the retained MME historical signaling data records are used as the basic data in the offline fingerprint database.

[0079] In this embodiment of the application, by using the location dataset of the second data source to correct the low-precision MME signaling data, a higher-precision offline fingerprint database base data is obtained, which improves the data reliability and accuracy of the offline fingerprint database and further improves the accuracy of outdoor positioning.

[0080] In one embodiment, the recorded data in the offline fingerprint database is obtained by feature derivation based on the basic data according to a preset feature dimension;

[0081] The preset feature dimensions include one or more of the following: weekday identifier, week identifier, hourly granularity identifier, first time window identifier, and historical motion trajectory identifier.

[0082] In this application embodiment, feature derivation can be performed on the fingerprint database basic data of the above embodiments. For example, one or more feature dimensions can be derived from each data record, such as weekday identifier, week identifier, hourly granularity identifier, first time window identifier, and historical movement trajectory identifier, thereby providing more dimensions of calculation and comparison for subsequent location information matching, and further improving the accuracy of outdoor positioning.

[0083] Please see Figure 2-4 Based on the above scheme, and to facilitate a better understanding of the outdoor positioning method provided in the embodiments of this application, the following detailed explanation is provided:

[0084] Based on the problems of the three existing outdoor positioning methods mentioned in the background art, this application proposes a real-time outdoor fingerprint positioning method based on multi-source data fusion. The main measures include: 1) real-time positioning based on MME signaling data to achieve near real-time and high coverage; 2) fusing multi-source high-precision location data and correcting some low-precision MME signaling data; 3) in-depth mining of MME user signaling location data to extract relevant features affecting the user's accurate location for constructing an offline fingerprint database; 4) in the real-time positioning stage, a multi-level filtering mechanism is proposed for preliminary pruning to improve the efficiency of the query algorithm; 5) based on the specificity of different features, corresponding similarity calculation methods are adopted to improve the positioning accuracy of the algorithm.

[0085] It should be noted that, since existing technologies cannot achieve near real-time and high coverage when building fingerprint databases based on MR data, while MME signaling data has a wide coverage and high collection rate, this application embodiment mines effective features from MME signaling data to build a fingerprint database and achieve real-time high-precision positioning. The overall process is as follows: Figure 2 As shown.

[0086] The first step is to build an offline fingerprint database. This process mainly consists of two steps: correcting low-precision MME signaling data and extracting fingerprint features. The specific steps are as follows.

[0087] 1.1 Correction of low-precision data in MME signaling.

[0088] Because the location data in MME signaling data (MME signaling data is data reported during user-base station interactions, reported when a user makes a call, accesses the internet, or switches base stations) contains low-precision base station locations, an effective fingerprint database cannot be built solely based on MME signaling data. MDT (Minimization of Drive-Test, a technology for automated collection and analysis of UE measurement reports containing location information in communication systems, used to minimize manual drive testing; it can be used to measure the distance between users and base stations, but has no azimuth angle, suffers from significant interference, has a latency of 4-5 hours, and an accuracy within 10 meters) and AlaDing (AlaDing data: AlaDing is a mobile terminal system where account managers report precise GPS location data) contain high-precision user locations, but the collection rate is low. Therefore, these two types of data can be correlated using user ID and time windows to correct the accuracy of some MME signaling data, and this data can be used as the basis for the offline fingerprint database. The specific correlation steps are as follows:

[0089] 1) First, rasterize the precise locations in the MDT and Ala-Ding;

[0090] 2) Each record in the MME signaling data has a start time and an end time. When the user ID in the MDT and Aladdin data matches the user ID in the MME data record, and the time of the MDT and Aladdin records is within the MME time window, they are associated and matched.

[0091] 3) If multiple precise location data can match the current MME record, the raster with the highest frequency will be used as the precise location raster for the current MME record; otherwise, no association will be made.

[0092] 4) Calculate the distance from the center point of the precise location grid to the MME base station of the MME record. If the distance exceeds a set threshold, delete the associated record. The final corrected data is the base data for the offline fingerprint database.

[0093] 1.2 Fingerprint database feature extraction.

[0094] It is understandable that most users' movement trajectories exhibit certain regularities, and the paths users take to certain locations are often similar, such as a user's daily commute route. Based on this, the embodiments of this application derive the following features from the fingerprint database basic data obtained in the first step:

[0095] 1) Time-related characteristics: Whether it is a working day;

[0096] 2) Derivation of time-related features: Week identifier;

[0097] 3) Time feature dimension derivation: hourly granularity identifier (which can be generalized to time periods such as morning, noon, and afternoon);

[0098] 4) Derivation of time feature dimension: Time window identifier (N minutes is a time window);

[0099] 5) Derivation of historical motion trajectory feature dimensions: Since the user trajectory of MME signaling data is low-precision data at the base station level, in order to effectively mine the historical regularity of user motion trajectory, it can be generalized by rasterizing the location of trajectory points, thereby deriving the location raster ID at time t-1, location raster ID at time t-2, ..., location raster ID at time tn.

[0100] After the above two steps, the final offline fingerprint database will be generated, in the following format:

[0101]

[0102] The second step is real-time positioning. After the offline fingerprint database is built, accurate positioning can be performed based on real-time MME signaling data. To improve the efficiency and accuracy of the positioning and matching algorithm, this application optimizes the process from two aspects: pruning and similarity measurement. The overall real-time positioning process is as follows: Figure 3 As shown.

[0103] 2.1 Pruning Search Based on Multi-Level Filtering Mechanism

[0104] In the real-time positioning phase, the baseline method calculates the similarity between real-time data features and each data feature in the offline fingerprint database through brute-force retrieval. However, due to the large size of the offline fingerprint database, this method incurs significant computational overhead and is inefficient, failing to meet the real-time requirements of outdoor positioning.

[0105] This application embodiment utilizes a multi-level filtering mechanism for initial pruning to generate a minimum subset of fingerprints that meets the conditions. During real-time localization, the algorithm only traverses and queries this minimum subset, significantly improving efficiency. As an example, the specific steps are as follows:

[0106] Assume the real-time dataset is A and the offline fingerprint database dataset is B. First, determine whether the offline fingerprint database dataset B contains the base station ID from dataset A. If it does, proceed to the fingerprint database matching stage; otherwise, directly output the latitude and longitude of the MME base station as the outdoor positioning information.

[0107] If the fingerprint database matching stage is reached, multi-level pruning can be performed. The first level: Filter records in dataset B that have the same base station ID as those in dataset A, denoted as b1;

[0108] Second layer: Filter out records in dataset b1 that have the same user ID as those in dataset A, and denote them as b2. If b2 is an empty set, then b2 = b1.

[0109] The third layer: In dataset b2, select records with the same weekday label as those in dataset A, and denote them as b3. If b3 is an empty set, then b3 = b2.

[0110] Fourth layer: In dataset b3, select records that have the same hourly granularity label as (or within one hour) as the hourly granularity label in dataset A, and denote them as b4. If b4 is an empty set, then b4 = b3.

[0111] Fifth layer: In dataset b4, select records that have the same time window label as (or differ by one time window) from dataset A, and denote them as b5. If b5 is an empty set, then b5 = b4; finally, b5 is the minimum subset of the fingerprint database that meets the conditions we need.

[0112] 2.2 Improvement of Similarity Calculation Method

[0113] After obtaining the minimum fingerprint subset, it is necessary to calculate the similarity between the data to be located and each data in the minimum fingerprint subset. Regarding similarity measurement, existing similarity measurement formulas are not applicable due to the specificity of each feature. This application proposes different similarity calculation methods based on the differences in the value range distribution of different features, making the calculation of feature similarity more reasonable. Furthermore, to improve the robustness of the real-time positioning algorithm, a weighted average is applied to the precise locations of the top k records with the highest final similarity in the offline fingerprint database to obtain the precise location prediction value. The final similarity can be measured from the following five dimensions.

[0114] 1) Distance of overlap between historical trajectory points

[0115]

[0116] Assume the first five location grid cells corresponding to the data to be located are R1 = [(2,1),(2,2),(3,3),(4,4),(5,5)]

[0117] The first five grid positions corresponding to a record in the smallest fingerprint subset are R2 = [(2,4),(2,3),(3,4),(3,6),(5,7)].

[0118] Next, each grid cell in R2 is expanded into a 3x3 grid. The expansion of grid (2,4) is as follows: Figure 4 As shown. Following this pattern, the 5 grid positions in R2 are eventually expanded to 45 grid positions. Next, we determine how many grid positions in R1 fall within these 45 grid positions, denoted as len_same; where len_all = 5, indicating that there are currently five comparison points.

[0119] 2) Distance of weekday identifier dimension:

[0120]

[0121] Where i wd To determine whether the target record is a workday, j wd To determine whether the records in the fingerprint database are from working days.

[0122] 3) Weekly identifier dimension distance:

[0123]

[0124] Where i wk To record the week identifier for the target, j wk This serves as a record cycle identifier in the fingerprint database.

[0125] 4) Hourly marker dimensional distance:

[0126]

[0127] Where i h To record the hour identifier for the target, j h This is the hour identifier for records in the fingerprint database.

[0128] 5) Time window identifier dimension distance:

[0129]

[0130]

[0131] Where i tw j is the identifier for the target recording time window. tw This serves as a time window identifier for records in the fingerprint database. size This represents the size of the time window.

[0132] The final distance is the sum of all distances; the smaller the distance value, the higher the similarity.

[0133] dis = dissame +dis workday +dis weekday +dis hour +dis tw

[0134] It should be noted that the key points of the embodiments of this application are as follows:

[0135] First, by integrating multiple high-precision data sources such as MDT and Aladdin, features are mined from the MME signaling data to construct a fingerprint database, thereby achieving near real-time and high coverage.

[0136] Second, based on business scenarios, a multi-level filtering mechanism was added to the matching algorithm as an initial pruning, which reduced the number of fingerprint database sets for similarity calculation and improved algorithm efficiency.

[0137] Third, unlike traditional similarity measurement methods, this paper adopts corresponding similarity measurement methods based on the various characteristics of data records, and redefines similarity from multiple dimensions such as historical trajectory overlap and time overlap.

[0138] Compared with the prior art, the embodiments of this application have the following beneficial effects:

[0139] 1. A fingerprint database is constructed by extracting features from MME signaling data, which achieves near real-time performance and high coverage compared to the method of constructing a fingerprint database based on MR data.

[0140] 2. Traditional methods employ brute-force searches across the entire offline fingerprint database, resulting in low efficiency. This application utilizes a multi-level filtering mechanism, prioritizing the pruning of the fingerprint database based on business characteristics to narrow the search scope and improve query efficiency.

[0141] 3. Traditional similarity measurement methods are not applicable to the fingerprint database features of this application. Therefore, this application adopts a new similarity calculation method. Based on the difference in the value range distribution of different features, different similarity calculation methods are formulated to make the calculation of feature similarity more reasonable.

[0142] 4. Based on the above multi-dimensional improvements, the outdoor fingerprint positioning model proposed in this application is highly efficient and accurate in practical applications, and the verification shows that the average error is within 60m.

[0143] The outdoor positioning device provided in the embodiments of this application is described below. The outdoor positioning device described below and the outdoor positioning method described above can be referred to in correspondence.

[0144] Please see Figure 5 This application provides an outdoor positioning device, which may include:

[0145] Data filtering module 1 is used to match the smallest subset of the fingerprint database from the pre-stored offline fingerprint database based on the real-time signaling data of the MME reported by the terminal under test.

[0146] Data calculation module 2 is used to calculate the similarity between the MME real-time signaling data and each record data in the minimum fingerprint subset;

[0147] The positioning determination module 3 is used to filter out several target record data with the highest similarity, and determine the real-time positioning information of the terminal under test based on the location information of the several target record data.

[0148] The basic data in the offline fingerprint database is generated by associating the MME historical signaling dataset with the location dataset in the second data source.

[0149] In one embodiment, the data filtering module 1 is specifically used for:

[0150] Based on the MME real-time signaling data reported by the terminal under test, the offline fingerprint database is pruned at least two levels to generate a minimum subset of the fingerprint database that matches the MME real-time signaling data.

[0151] In one embodiment, the data calculation module 2 is specifically used for:

[0152] For each record data in the minimum fingerprint subset, calculate the dimensional distance between the MME real-time signaling data and the record data in at least two feature dimensions;

[0153] The similarity between the MME real-time signaling data and the recorded data is determined based on the dimensional distance on at least two feature dimensions.

[0154] In one embodiment, the positioning determination module 3 is specifically used for:

[0155] Filter out the target record data with the highest similarity and obtain the location information corresponding to the target record data;

[0156] Based on the similarity between each of the target record data and the MME real-time signaling data, a weighted average of the several location information data is performed to determine the real-time location information of the terminal under test.

[0157] In one embodiment, the basic data in the offline fingerprint database is generated by associating the MME historical signaling dataset with the location dataset in the second data source, including:

[0158] Rasterize the location dataset in the second data source;

[0159] The location dataset and the MME historical signaling dataset are matched according to the user ID and time window, and the matching MME historical signaling data records are associated with the corresponding location data records.

[0160] The location raster of the location data record that appears most frequently is used as the precise location raster of the corresponding MME historical signaling data record;

[0161] MME historical signaling data records where the distance from the center point of the precise location grid to the MME base station exceeds a preset threshold are deleted, and the retained MME historical signaling data records are used as the basic data in the offline fingerprint database.

[0162] In one embodiment, the recorded data in the offline fingerprint database is obtained by feature derivation based on the basic data according to a preset feature dimension;

[0163] The preset feature dimensions include one or more of the following: weekday identifier, week identifier, hourly granularity identifier, first time window identifier, and historical motion trajectory identifier.

[0164] In one embodiment, the pruning process on the offline fingerprint database at least two levels includes at least two of the following:

[0165] The offline fingerprint database records that do not match the base station ID of the MME real-time signaling data are filtered out;

[0166] The offline fingerprint database records that do not match the user IDs in the MME real-time signaling data are filtered out.

[0167] The record data in the offline fingerprint database that does not match the weekday label of the MME real-time signaling data is filtered out;

[0168] The offline fingerprint database records that do not match the hourly granularity labels of the MME real-time signaling data are filtered out.

[0169] Records in the offline fingerprint database that do not match the time window labels of the MME real-time signaling data are filtered out.

[0170] It is understood that the above-described device embodiments correspond to the method embodiments of this application. The outdoor positioning device provided in the embodiments of this application can implement the outdoor positioning method provided in any one of the method embodiments of this application.

[0171] Figure 6 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 6As shown, the electronic device may include: a processor 610, a communication interface 620, a memory 630, and a communication bus 640, wherein the processor 610, the communication interface 620, and the memory 630 communicate with each other via the communication bus 640. The processor 610 can call a computer program in the memory 630 to execute an outdoor positioning method, such as including:

[0172] S1. Based on the MME real-time signaling data reported by the terminal under test, match the smallest subset of the fingerprint database from the pre-stored offline fingerprint database.

[0173] S2. Calculate the similarity between the MME real-time signaling data and each record data in the minimum fingerprint subset;

[0174] S3. Select several target record data with the highest similarity, and determine the real-time location information of the terminal under test based on the location information of the several target record data.

[0175] The basic data in the offline fingerprint database is generated by associating the MME historical signaling dataset with the location dataset in the second data source.

[0176] Furthermore, the logical instructions in the aforementioned memory 630 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0177] On the other hand, embodiments of this application also provide a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can perform the outdoor positioning methods provided in the above embodiments, such as including:

[0178] S1. Based on the MME real-time signaling data reported by the terminal under test, match the smallest subset of the fingerprint database from the pre-stored offline fingerprint database.

[0179] S2. Calculate the similarity between the MME real-time signaling data and each record data in the minimum fingerprint subset;

[0180] S3. Select several target record data with the highest similarity, and determine the real-time location information of the terminal under test based on the location information of the several target record data.

[0181] The basic data in the offline fingerprint database is generated by associating the MME historical signaling dataset with the location dataset in the second data source.

[0182] On the other hand, embodiments of this application also provide a processor-readable storage medium storing a computer program for causing a processor to execute the methods provided in the above embodiments, such as including:

[0183] S1. Based on the MME real-time signaling data reported by the terminal under test, match the smallest subset of the fingerprint database from the pre-stored offline fingerprint database.

[0184] S2. Calculate the similarity between the MME real-time signaling data and each record data in the minimum fingerprint subset;

[0185] S3. Select several target record data with the highest similarity, and determine the real-time location information of the terminal under test based on the location information of the several target record data.

[0186] The basic data in the offline fingerprint database is generated by associating the MME historical signaling dataset with the location dataset in the second data source.

[0187] The processor-readable storage medium can be any available medium or data storage device that the processor can access, including but not limited to magnetic memory (e.g., floppy disk, hard disk, magnetic tape, magneto-optical disk (MO)), optical memory (e.g., CD, DVD, BD, HVD), and semiconductor memory (e.g., ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid-state drive (SSD)).

[0188] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0189] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0190] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. An outdoor positioning method, characterized in that, include: Based on the real-time signaling data of the MME reported by the terminal under test, the smallest subset of the fingerprint database is matched from the pre-stored offline fingerprint database. Calculate the similarity between the MME real-time signaling data and each record data in the minimum fingerprint subset, wherein the similarity is determined based on the final distance between the MME real-time signaling data and the record data, and the final distance is the sum of the historical trajectory point overlap distance, the weekday identifier dimension distance, the week identifier dimension distance, the hour identifier dimension distance, and the time window identifier dimension distance; The system filters out several target record data with the highest similarity and determines the real-time location information of the terminal under test based on the location information of the several target record data. The basic data in the offline fingerprint database is generated by associating the MME historical signaling dataset with the location dataset in the second data source.

2. The outdoor positioning method according to claim 1, characterized in that, The step of matching the minimum subset of fingerprints from the pre-stored offline fingerprint database based on the MME real-time signaling data reported by the terminal under test includes: Based on the MME real-time signaling data reported by the terminal under test, the offline fingerprint database is pruned at least two levels to generate a minimum subset of the fingerprint database that matches the MME real-time signaling data.

3. The outdoor positioning method according to claim 1, characterized in that, The calculation of the similarity between the MME real-time signaling data and each record in the minimum fingerprint subset includes: For each record data in the minimum fingerprint subset, calculate the dimensional distance between the MME real-time signaling data and the record data in at least two feature dimensions; The similarity between the MME real-time signaling data and the recorded data is determined based on the dimensional distance on at least two feature dimensions.

4. The outdoor positioning method according to claim 1, characterized in that, The process of selecting several target record data with the highest similarity and determining the real-time location information of the terminal under test based on the location information of the several target record data includes: Filter out the target record data with the highest similarity and obtain the location information corresponding to the target record data; Based on the similarity between each of the target record data and the MME real-time signaling data, a weighted average of the several location information data is performed to determine the real-time location information of the terminal under test.

5. The outdoor positioning method according to claim 1, characterized in that, The basic data in the offline fingerprint database is generated by associating the MME historical signaling dataset with the location dataset in the second data source, including: Rasterize the location dataset in the second data source; The location dataset and the MME historical signaling dataset are matched according to the user ID and time window, and the matching MME historical signaling data records are associated with the corresponding location data records. The location raster of the location data record that appears most frequently is used as the precise location raster of the corresponding MME historical signaling data record; MME historical signaling data records where the distance from the center point of the precise location grid to the MME base station exceeds a preset threshold are deleted, and the retained MME historical signaling data records are used as the basic data in the offline fingerprint database.

6. The outdoor positioning method according to claim 1, characterized in that, The recorded data in the offline fingerprint database is derived from the basic data according to a preset feature dimension. The preset feature dimensions include one or more of the following: weekday identifier, week identifier, hourly granularity identifier, first time window identifier, and historical motion trajectory identifier.

7. The outdoor positioning method according to claim 2, characterized in that, The pruning process on the offline fingerprint database, consisting of at least two levels, includes at least two of the following: The offline fingerprint database records that do not match the base station ID of the MME real-time signaling data are filtered out; The offline fingerprint database records that do not match the user IDs in the MME real-time signaling data are filtered out. The record data in the offline fingerprint database that does not match the weekday label of the MME real-time signaling data is filtered out; The offline fingerprint database records that do not match the hourly granularity labels of the MME real-time signaling data are filtered out. Records in the offline fingerprint database that do not match the time window labels of the MME real-time signaling data are filtered out.

8. An outdoor positioning device, characterized in that, include: The data filtering module is used to match the smallest subset of the fingerprint database from the pre-stored offline fingerprint database based on the real-time signaling data of the MME reported by the terminal under test. The data calculation module is used to calculate the similarity between the MME real-time signaling data and each record data in the minimum fingerprint subset, wherein the similarity is determined based on the final distance between the MME real-time signaling data and the record data, and the final distance is the sum of the historical trajectory point overlap distance, the weekday identifier dimension distance, the week identifier dimension distance, the hour identifier dimension distance, and the time window identifier dimension distance; The positioning determination module is used to filter out several target record data with the highest similarity, and determine the real-time positioning information of the terminal under test based on the location information of the several target record data. The basic data in the offline fingerprint database is generated by associating the MME historical signaling dataset with the location dataset in the second data source.

9. An electronic device comprising a processor and a memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the outdoor positioning method according to any one of claims 1 to 7.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the outdoor positioning method according to any one of claims 1 to 7.

Images