Mobile Network User Location Method and System Based on MRO and DPI Data Correlation

Abstract

This invention discloses a mobile network user positioning method and system based on the association of MRO and DPI data, belonging to the field of mobile network positioning technology. The technical problem this invention aims to solve is how to improve the accuracy of user location positioning while reducing construction costs. The technical solution is as follows: The method specifically involves: obtaining user latitude and longitude information (OTT positioning) through user plane DPI data; establishing a fingerprint database using the user latitude and longitude extracted from OTT positioning; backfilling user information from MRO data using signaling plane DPI data; and finally establishing the association between MRO data and the OTT fingerprint database; user positioning based on MR_TA+AoA or three-point positioning based on MR_TDOA: using parameter information in the MRO data, locating the user's latitude and longitude using the TA+AoA or TDOA three-point positioning method; and MRO positioning result optimization based on the fingerprint database: effectively associating the MRO positioning results with the OTT fingerprint database to optimize and improve the accuracy of the MRO positioning results. The system includes a fingerprint database establishment unit, a positioning unit, and an optimization unit.

Description

Technical Field

[0001] This invention relates to the field of mobile network positioning technology, specifically a method and system for locating mobile network users based on the association of MRO and DPI data. Background Technology

[0002] With the rapid development of the mobile internet, user location tracking needs are becoming increasingly diverse: in user safety, such as locating lost elderly / children and forest firefighters; in location-based marketing, such as targeted advertising and regional marketing; and in security, such as user trajectory tracking and setting up electronic fences for special users, there are demands in multiple fields. The development speed of mobile communication networks far surpasses that of desktop internet. Due to the massive number of mobile network users, obtaining mobile terminal location information to realize various location-related value-added services for people and objects has become a hot topic in internet commerce.

[0003] Mobile user positioning technologies typically rely on three main types of networks: satellite positioning, WLAN positioning, and mobile network positioning. Satellite positioning is suitable for outdoor scenarios, while WLAN positioning is suitable for indoor scenarios. Both of these methods require terminal modifications, resulting in a certain level of complexity and construction cost. Mobile network positioning, on the other hand, is applicable to both indoor and outdoor scenarios, and its positioning accuracy is continuously improving with large-scale network deployment.

[0004] With the development of mobile positioning technology, various mobile user positioning solutions have emerged, such as AGPS positioning, TOA&TDOA positioning, and TA+AoA positioning. Although AGPS positioning has high accuracy, it relies on GPS hardware modules, which is costly. TOA&TDOA positioning obtains the position of the mobile station by measuring the time difference, which requires the installation of additional hardware such as measurement units in the base station.

[0005] Through comparison of domestic and international mobile network positioning technologies, and research into the background of mobile user positioning technologies, it was found that in the MRO (Measurement Recovery Report) on the wireless network side, signal propagation between the mobile station and the base station carries data such as propagation time, signal strength, and direction angle. This information can be used comprehensively to achieve location positioning. However, MRO data lacks unique user information, making it impossible to identify individual users. Analysis of DPI (Dedicated Point Indicator) data revealed that DPI user plane data, such as the S1-U / N3 interface, contains user latitude and longitude coordinates, sourced from location information reported by web / app applications. The accuracy of user latitude and longitude in DPI data can reach approximately 50 meters indoors, and 5-10 meters or even lower outdoors. However, the latitude and longitude coordinates in DPI data are only available to users using certain web / app applications, and cannot be used to achieve location positioning for all users across the network.

[0006] Therefore, how to reduce construction costs while improving the accuracy of user location positioning is a technical problem that urgently needs to be solved. Summary of the Invention

[0007] The technical objective of this invention is to provide a mobile network user positioning method and system based on the association of MRO and DPI data, in order to solve the problem of how to reduce construction costs while improving the accuracy of user location positioning.

[0008] The technical objective of this invention is achieved as follows: a mobile network user location method based on the association of MRO and DPI data, the method being as follows:

[0009] Establish a fingerprint database: Obtain user latitude and longitude information, i.e., OTT positioning, through user plane DPI data (latitude and longitude information carrying ratio is about 1-3%), and establish a fingerprint database using the user latitude and longitude extracted by OTT positioning; at the same time, use signaling plane DPI data to realize the backfilling of user information in MRO data, and finally establish the association between MRO data and OTT fingerprint database;

[0010] User positioning based on MR_TA+AoA or three-point positioning based on MR_TDOA: Using parameter information in MRO data, the latitude and longitude of the user are located through the TA+AoA or TDOA three-point positioning method;

[0011] MRO location result optimization based on fingerprint database: Effectively associate MRO location results with OTT fingerprint database to optimize and fine-tune MRO location results, improve the accuracy of MRO location results, and achieve the goal of accurate location of users across the network.

[0012] As a preferred option, the fingerprint database is established as follows:

[0013] Obtaining user latitude and longitude based on DPI data: Connect to the existing network signaling monitoring system or collect user S1-U interface to obtain user latitude and longitude information from HTTP protocol call detail records;

[0014] Linking DPI data and MRO data: Using signaling plane DPI data to backfill user information in MRO data, and then linking the backfilled MRO with the S1-U interface to obtain the network coverage level and quality value of the user at any latitude and longitude, as well as information such as distance and azimuth from the base station.

[0015] Fingerprint database based on DPI_OTT data: A supervised machine learning model is used to build a fingerprint feature model based on the latitude and longitude data of OTT users.

[0016] More optimally, the user's latitude and longitude can be obtained by relying on DPI data as follows:

[0017] When users use certain apps for data services, they will exchange information with the server via the S1-U interface through the HTTP protocol. Some of this information includes the user's real latitude and longitude information.

[0018] By collecting and analyzing network data using DPI technology, the latitude and longitude information of a certain percentage of users across the entire network can be obtained.

[0019] Better still, a supervised machine learning model is used to build a fingerprint feature model based on the latitude and longitude data of OTT users, as follows:

[0020] The user's latitude and longitude data obtained via OTT are distributed in a 10*10 grid.

[0021] Feature extraction is performed on information such as serving cell ID, serving cell RSRP / RSRQ, Tadv, AoA, neighboring cell ID and neighboring cell RSRP / RSRQ, which serve as the feature basis for machine learning models.

[0022] Using a large amount of OTT latitude and longitude reported data from the existing network, we trained a machine learning model with fingerprint features.

[0023] Based on the results of feature analysis, a fingerprint database is generated according to the feature identifiers of grids, main cells, and neighboring cells;

[0024] Establish a fingerprint database update mechanism: continuously perform machine learning through updates to OTT data to improve the accuracy of the fingerprint database.

[0025] As a preferred method, user location based on MR_TA+AoA is as follows:

[0026] The MRO data already contains TA+AoA parameter information. Combining AoA and TA, UE positioning is performed based on a single cell. The formula for the TA+AoA positioning principle is as follows:

[0027]

[0028] Where TA represents the time lead; c represents the speed of light, and the value of c is 3.0 * 10^- ... 8 m / s;

[0029] The time advance distance corresponding to 1 Ts is: (3*108*1 / (15000*2048)) / 2=4.89m, which means that the distance = propagation speed (speed of light) * 1Ts / 2 (uplink and downlink path sum); the TA value reported by MR is in 16TS, 1TADV=16TS=16*4.89=78.12m; the distance from the terminal UE to the antenna d=78.12*TA, in meters;

[0030] The distance calculated based on TA or path loss is the distance from the UE to the antenna port, which is a three-dimensional distance with an elevation angle. Normally, the UE's altitude is lower than the eNB's altitude, while the user's latitude and longitude change in two dimensions. The specific calculation of the distance from the UE to the base station is as follows:

[0031] Ignoring the UE height, according to the Pythagorean theorem: L 2 +H 2 =d 2 The straight-line distance L from the UE to the antenna and the base station height H are derived from the engineering parameter base station height.

[0032] The latitude and longitude information of the base station is obtained by obtaining the engineering parameters, and the latitude and longitude (X0, Y0) of the terminal user can be obtained by conversion.

[0033] As a preferred method, the three-point localization based on MR_TDOA is as follows:

[0034] The three-point positioning algorithm requires that the effective number of base stations for calculating the "distance from terminal to base station" is 3, which are the level values ​​of the primary serving cell and at least two neighboring cells extracted from the MR data. Using the wireless propagation model algorithm, the distance d1 from the terminal to the primary serving cell and the distances d2 and d3 from the neighboring base stations are obtained respectively.

[0035] The location of the terminal user from the base station and neighboring base stations is obtained using the empirical formula for 4G path loss calculation. The empirical formula for 4G path loss calculation is as follows:

[0036] L COST231-Hata =46.3 + 33.9 * log 10 (f c -13.82*log 10 (h b )+(44.9-6.55*log 10 (h)*log 10 (d)+C M

[0037] Among them, f c The wireless signal frequency is 1500-2000MHz, in MHz; c M To cover the scene correction factor, cover_class covers rural areas, towns, general urban areas, and core urban areas. M The values ​​are 0, 3, and 6 respectively; h is the height difference between the terminal and the base station antenna, which is equal to the base station height (transmitting antenna height) in the engineering parameters by default, in meters; d is the distance between the base station antenna and the mobile station antenna (antenna coverage distance), in kilometers.

[0038] By using the least squares method combined with the latitude and longitude information of the base station, the latitude and longitude (X1, Y1) of the terminal user can be obtained after conversion.

[0039] As a preferred option, the MRO localization results based on the fingerprint database are optimized as follows:

[0040] Based on MRO data, the TA+AoA positioning algorithm is affected by multipath propagation and interference from coverage scene, resulting in the latitude and longitude accuracy of user positioning ranging from 50 to 1000m. The OTT positioning fingerprint database model is effectively utilized. By minimizing the Euclidean distance algorithm, the MRO positioning result is mapped to the fingerprint database grid. The grid whose feature information is closest to the feature information contained in the current MRO positioning result is found. Finally, the position of this grid is used as the position of the corresponding terminal user.

[0041] Simultaneously, the TA+AOA or TDOA three-point positioning results data under different coverage scenarios (urban / rural / universities / office buildings, etc.) are continuously input into the OTT fingerprint database to build a machine learning model, calculate the offset parameters of MR positioning results under different coverage scenarios, so as to further improve the accuracy of MR positioning results.

[0042] A mobile network user positioning system based on the association of MRO and DPI data, the system comprising,

[0043] The fingerprint database establishment unit is used to obtain the user's latitude and longitude, i.e., OTT positioning, through user plane DPI data (latitude and longitude information carrying ratio is about 1-3%), establish the fingerprint database using the user's latitude and longitude extracted from OTT positioning, and use signaling plane DPI data to realize the backfilling of user information in MRO data, and finally establish the association between MRO data and OTT fingerprint database.

[0044] The positioning unit is used to locate the user's latitude and longitude using parameter information in MRO data through the TA+AoA or TDOA three-point positioning method;

[0045] The optimization unit is used to effectively associate MRO positioning results with the OTT fingerprint database, optimize and fine-tune the MRO positioning results, improve the accuracy of MRO positioning results, and achieve the goal of accurate location positioning of users across the entire network.

[0046] An electronic device includes: a memory and at least one processor;

[0047] The memory contains computer programs;

[0048] The at least one processor executes the computer program stored in the memory, causing the at least one processor to perform the mobile network user location method based on the association of MRO and DPI data as described above.

[0049] A computer-readable storage medium storing a computer program that can be executed by a processor to implement the mobile network user location method based on the association of MRO and DPI data as described above.

[0050] The mobile network user positioning method and system based on the association of MRO and DPI data of the present invention have the following advantages:

[0051] (I) This invention achieves user identity backfilling of MRO data by closely integrating wireless MRO data with core network DPI data. At the same time, it adopts a variety of user location positioning methods to scientifically optimize the positioning results to reduce user location offset and output accurate user latitude and longitude information.

[0052] (II) This invention effectively combines MRO positioning results with DPI data and comprehensively adopts multiple positioning methods such as OTT positioning, TA+AoA positioning, and TDOA three-point positioning to achieve the location positioning of users across the network in a more cost-effective and efficient way.

[0053] (III) This invention effectively combines MRO data with DPI data, integrates the location service provided by the Internet in DPI data (i.e. OTT positioning) with the positioning TA+AOA or TDOA positioning of wireless MRO data, and forms a high-precision, low-cost user location positioning method.

[0054] (iv) This invention effectively combines MRO data with DPI data, thereby reducing construction costs while improving the accuracy of user location positioning.

[0055] (V) The advantages of the user positioning scheme based on the association of MRO and DPI data in this invention are:

[0056] ① OTT positioning based on DPI data has high accuracy, but it can only be used when the user is using certain apps, which cannot fully meet the user positioning needs of LBS. However, if a fingerprint database is built based on OTT positioning data and then associated with the positioning results of MRO (TA+AOA / TDOA three-point positioning), the problem of low positioning coverage can be effectively solved, and the positioning accuracy can also be effectively improved.

[0057] ② In principle, no additional hardware configuration is required, the production system will not be affected, and the construction cost is low. MRO data is generated by the base station starting the measurement configuration and reporting the measurement report to the NMS management system. MRO data can be synchronized from the NMS system. DPI data is synchronized by the existing network signaling monitoring system. In principle, no new collection nodes are required. Attached Figure Description

[0058] The invention will be further described below with reference to the accompanying drawings.

[0059] Appendix Figure 1 A flowchart of a mobile network user location method based on the association of MRO and DPI data;

[0060] Appendix Figure 2 A schematic diagram illustrating the correlation between DPI data and MR0 data;

[0061] Appendix Figure 3 A schematic diagram illustrating the establishment of a fingerprint database based on DPI_OTT data;

[0062] Appendix Figure 4 TA+AoA positioning principle diagram;

[0063] Appendix Figure 5 This diagram illustrates the calculation of the distance from the UE to the base station. Detailed Implementation

[0064] The mobile network user positioning method and system based on MRO and DPI data association of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0065] Example 1:

[0066] As attached Figure 1 As shown in the figure, this embodiment provides a mobile network user location method based on the association of MRO and DPI data. The method is as follows:

[0067] S1. Establish a fingerprint database: Obtain user latitude and longitude information, i.e., OTT positioning, through user plane DPI data (latitude and longitude information carrying ratio is about 1-3%), and establish a fingerprint database using the user latitude and longitude extracted by OTT positioning; at the same time, use signaling plane DPI data to realize the backfilling of user information in MRO data, and finally establish the association between MRO data and OTT fingerprint database.

[0068] S2. User positioning based on MR_TA+AoA or three-point positioning based on MR_TDOA: Using parameter information in MRO data, the latitude and longitude of the user are located by the TA+AoA or TDOA three-point positioning method.

[0069] S3. MRO Location Result Optimization Based on Fingerprint Database: Effectively associate MRO location results with OTT fingerprint database, optimize and fine-tune MRO location results, improve the accuracy of MRO location results, and achieve the goal of accurate location of users across the entire network.

[0070] The specific steps for establishing the fingerprint database in step S2 of this embodiment are as follows:

[0071] S201. Obtaining user latitude and longitude based on DPI data: Connect to the existing network signaling monitoring system or collect user S1-U interface to obtain user latitude and longitude information from HTTP protocol call detail records;

[0072] Correlation between S202, DPI data, and MR0 data: see attached. Figure 2 As shown, the MRO itself does not carry a user identifier; instead, user identity information is filled in for the MRO using the S1-MME interface. After the user identity information is filled in, the MRO is associated with the S1-U interface to obtain information such as the network coverage level and quality value of the user at a certain latitude and longitude, as well as the distance and azimuth of the user from the base station.

[0073] S203. Establish a fingerprint database based on DPI_OTT data: Use a supervised machine learning model to establish a fingerprint feature model based on the latitude and longitude data of OTT users.

[0074] The specific process by which the current network signaling monitoring system obtains the user's latitude and longitude location information is as follows:

[0075] Extracting location information from downlink 2000K messages: extracting latitude and longitude in HTML format from PAYLOAD and latitude and longitude in text format from PAYLOAD;

[0076] Extracting location information from upstream messages: Extracting latitude and longitude information from the URL;

[0077] The location information is extracted from the downlink 2000K messages and the uplink messages, and then the coordinate system is transformed to determine the user's latitude and longitude location information.

[0078] In this embodiment, step S201, which involves obtaining the user's latitude and longitude based on DPI data, is as follows:

[0079] S20101. When users use certain apps for data services, they will exchange information with the server through the S1-U interface via the HTTP protocol. Some of the information contains the user's real latitude and longitude information.

[0080] S20102. By collecting and analyzing network data using DPI technology, the latitude and longitude information of a certain proportion of users across the entire network can be obtained.

[0081] As attached Figure 3 As shown, in step S203 of this embodiment, the use of a supervised machine learning model to establish a fingerprint feature model based on the location latitude and longitude data of OTT users is as follows:

[0082] S20301. Distribute the user's latitude and longitude data obtained via OTT within a 10*10 grid.

[0083] S20302. Based on information such as serving cell ID, serving cell RSRP / RSRQ, Tadv, AoA, neighboring cell ID and neighboring cell RSRP / RSRQ, feature extraction is performed as the feature basis for the machine learning model.

[0084] S20303. Utilize a large amount of OTT latitude and longitude reporting data from the existing network to train the fingerprint features of the machine learning model;

[0085] S20304. Based on the results of feature analysis, generate a fingerprint database according to the feature identifiers of grids, main cells, and neighboring cells;

[0086] S20305. Establish a fingerprint database update mechanism: continuously perform machine learning through updates to OTT data to improve the accuracy of the fingerprint database.

[0087] The specific steps of user localization based on MR_TA+AoA in step S2 of this embodiment are as follows:

[0088] The MRO data already contains TA+AoA parameter information. Combining AoA and TA, UE positioning is performed based on a single cell; as shown in the attached figure. Figure 4 As shown, the formula for the TA+AoA positioning principle is as follows:

[0089]

[0090] Where TA represents the time lead; c represents the speed of light, and the value of c is 3.0 * 10^- ... 8 m / s;

[0091] The time advance distance corresponding to 1 Ts is: (3*108*1 / (15000*2048)) / 2=4.89m, which means that the distance = propagation speed (speed of light) * 1Ts / 2 (uplink and downlink path sum); the TA value reported by MR is in 16TS, 1TADV=16TS=16*4.89=78.12m; the distance from the terminal UE to the antenna d=78.12*TA, in meters;

[0092] The distance calculated based on TA or path loss is the distance from the UE to the antenna port, which is a three-dimensional distance with an elevation angle. Normally, the UE's altitude is lower than the eNB's altitude, while the user's latitude and longitude change, resulting in a two-dimensional change, as shown in the attached diagram. Figure 5 As shown, the specific steps for calculating the distance from the UE to the base station are as follows:

[0093] Ignoring the UE height, according to the Pythagorean theorem: L 2 +H 2 =d 2 The straight-line distance L from the UE to the antenna and the base station height H are derived from the engineering parameter base station height.

[0094] The latitude and longitude information of the base station is obtained by obtaining the engineering parameters, and the latitude and longitude (X0, Y0) of the terminal user can be obtained by conversion.

[0095] The specific details of the MR_TDOA-based three-point localization in step S2 of this embodiment are as follows:

[0096] The three-point positioning algorithm requires that the effective number of base stations for calculating the "distance from terminal to base station" is 3; extract the level value of the primary serving cell and the level values ​​of at least two neighboring cells from the MR data, and use the wireless propagation model algorithm to obtain the distance d1 from the terminal to the primary serving cell and the distances d2 & d3 from the neighboring base stations, respectively.

[0097] The location of the terminal user from the base station and neighboring base stations is obtained using the empirical formula for 4G path loss calculation. The empirical formula for 4G path loss calculation is as follows:

[0098] L COST231-Hata =46.3 + 33.9 * log 10 (f c -13.82*log 10 (h b )+(44.9-6.55*log 10 (h)*log 10 (d)+C M

[0099] Among them, f c The wireless signal frequency is 1500-2000MHz, in MHz; c M To cover the scene correction factor, cover_class covers rural areas, towns, general urban areas, and core urban areas. M The values ​​are 0, 3, and 6 respectively; h is the height difference between the terminal and the base station antenna, which is equal to the base station height (transmitting antenna height) in the engineering parameters by default, in meters; d is the distance between the base station antenna and the mobile station antenna (antenna coverage distance), in kilometers.

[0100] By using the least squares method combined with the latitude and longitude information of the base station, the latitude and longitude (X1, Y1) of the terminal user can be obtained after conversion.

[0101] The specific optimization of MRO positioning results based on the fingerprint database in step S3 of this embodiment is as follows:

[0102] Based on MRO data, the TA+AoA positioning algorithm is affected by multipath propagation and interference from coverage scene, resulting in the latitude and longitude accuracy of user positioning ranging from 50 to 1000m. The OTT positioning fingerprint database model is effectively utilized. By minimizing the Euclidean distance algorithm, the MRO positioning result is mapped to the fingerprint database grid. The grid whose feature information is closest to the feature information contained in the current MRO positioning result is found. Finally, the position of this grid is used as the position of the corresponding terminal user.

[0103] Simultaneously, the TA+AOA or TDOA three-point positioning results data under different coverage scenarios (urban / rural / universities / office buildings, etc.) are continuously input into the OTT fingerprint database to build a machine learning model, calculate the offset parameters of MR positioning results under different coverage scenarios, so as to further improve the accuracy of MR positioning results.

[0104] Example 2:

[0105] This embodiment provides a mobile network user positioning system based on the association of MRO and DPI data. The system includes:

[0106] The fingerprint database establishment unit is used to obtain the user's latitude and longitude, i.e., OTT positioning, through user plane DPI data (latitude and longitude information carrying ratio is about 1-3%), establish the fingerprint database using the user's latitude and longitude extracted from OTT positioning, and use signaling plane DPI data to realize the backfilling of user information in MRO data, and finally establish the association between MRO data and OTT fingerprint database.

[0107] The positioning unit is used to locate the user's latitude and longitude using parameter information in MRO data through the TA+AoA or TDOA three-point positioning method;

[0108] The optimization unit is used to effectively associate MRO positioning results with the OTT fingerprint database, optimize and fine-tune the MRO positioning results, improve the accuracy of MRO positioning results, and achieve the goal of accurate location positioning of users across the entire network.

[0109] Example 3:

[0110] This embodiment also provides an electronic device, including: a memory and a processor;

[0111] The memory stores the instructions executed by the computer.

[0112] The processor executes computer execution instructions stored in the memory, causing the processor to execute the mobile network user location method based on MRO and DPI data association in any embodiment of the present invention.

[0113] The processor can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), off-the-shelf programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor can be a microprocessor or any conventional processor.

[0114] Memory is used to store computer programs and / or modules. The processor implements various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory, and by accessing data stored in the memory. Memory can mainly include a program storage area and a data storage area. The program storage area can store the operating system, at least one application program required for a function, etc.; the data storage area can store data created based on the use of the terminal, etc. In addition, memory can also include high-speed random access memory, and can also include non-volatile memory, such as hard disks, RAM, plug-in hard disks, smart memory cards (SMC), secure digital cards (SD cards), flash memory cards, at least one disk storage device, flash memory devices, or other volatile solid-state storage devices.

[0115] Example 4:

[0116] This embodiment also provides a computer-readable storage medium storing multiple instructions, which are loaded by a processor to cause the processor to execute the mobile network user positioning method based on MRO and DPI data association in any embodiment of the present invention. Specifically, a system or apparatus equipped with a storage medium may be provided, on which software program code implementing the functions of any of the above embodiments is stored, and the computer (or CPU or MPU) of the system or apparatus may read and execute the program code stored in the storage medium.

[0117] In this case, the program code read from the storage medium can itself implement the function of any of the above embodiments, and therefore the program code and the storage medium storing the program code constitute part of the present invention.

[0118] Storage media embodiments for providing program code include floppy disks, hard disks, magneto-optical disks, optical disks (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RYM, DVD-RW, DVD+RW), magnetic tapes, non-volatile memory cards, and ROMs. Alternatively, program code can be downloaded from a server computer via a communication network.

[0119] Furthermore, it should be clear that not only can the program code read by the computer be executed, but also the operating system or other components operating on the computer can be instructed based on the program code to perform some or all of the actual operations, thereby realizing the function of any of the embodiments described above.

[0120] Furthermore, it is understood that the program code read from the storage medium is written to the memory set in the expansion board inserted into the computer or to the memory set in the expansion unit connected to the computer. Then, based on the instructions of the program code, the CPU or other components installed on the expansion board or expansion unit execute some and all of the actual operations, thereby realizing the function of any of the embodiments described above.

[0121] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention 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 or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A mobile network user location method based on the association of MRO and DPI data, characterized in that, The method is as follows: Establish a fingerprint database: Obtain user latitude and longitude information, i.e., OTT positioning, through user plane DPI data, and use the user latitude and longitude extracted by OTT positioning to establish a fingerprint database; at the same time, use signaling plane DPI data to realize the backfilling of user information in MRO data, and finally establish the association between MRO data and OTT fingerprint database; User positioning based on MR_TA+AoA or three-point positioning based on MR_TDOA: Using parameter information in MRO data, the latitude and longitude of the user are located through the TA+AoA or TDOA three-point positioning method; MRO location result optimization based on fingerprint database: Effectively associate MRO location results with OTT fingerprint database to optimize and improve the accuracy of MRO location results; The establishment of the fingerprint database is specifically as follows: Obtaining user latitude and longitude based on DPI data: Connect to the existing network signaling monitoring system or collect user S1-U interface to obtain user latitude and longitude information from HTTP protocol call detail records; DPI data and MRO data are correlated: MRO itself does not carry user identifiers. User identity information is backfilled into MRO using the S1-MME interface. After the user identity information is backfilled, MRO is correlated with the S1-U interface to obtain the network coverage level and quality value of the user at any latitude and longitude, as well as the distance and azimuth information from the base station. Fingerprint database based on DPI_OTT data: A supervised machine learning model is used to build a fingerprint feature model based on the latitude and longitude data of OTT users. The specific steps for obtaining user latitude and longitude based on DPI data are as follows: When users use the APP for data services, they exchange information with the server through the S1-U interface via the HTTP protocol. Some of the information contains the user's real latitude and longitude information. By collecting and analyzing network data using DPI technology, the latitude and longitude information of the corresponding proportion of users across the entire network can be obtained; The fingerprint feature model is established using supervised machine learning model based on the latitude and longitude data of OTT users, as follows: The user latitude and longitude data obtained through OTT are distributed across 10 points. Within a grid of 10; Feature extraction is performed based on serving cell ID, serving cell RSRP / RSRQ, Tadv, AoA, neighboring cell ID, and neighboring cell RSRP / RSRQ information, which serve as the feature basis for the machine learning model. Using existing OTT latitude and longitude data, we trained a machine learning model with fingerprint features. Based on the results of feature analysis, a fingerprint database is generated according to the feature identifiers of grids, main cells, and neighboring cells; Establish a fingerprint database update mechanism: continuously perform machine learning through updates to OTT data to improve the accuracy of the fingerprint database; The specific user location based on MR_TA+AoA is as follows: The MRO data already contains TA+AoA parameter information. Combining AoA and TA, UE positioning is performed based on a single cell. The formula for the TA+AoA positioning principle is as follows: ； in, Indicates the lead time; Represents the speed of light. The value is 3.0 m / s; The time advance distance corresponding to 1Ts is: (3 10 8 1 / (15000 2048)) / 2=4.89m, which means that distance = speed of light. 1Ts / 2; MR reports TA values ​​in units of 16TS, 1TADV = 16TS = 16 4.89 = 78.12m; Distance from the terminal UE to the antenna d = 78.12m. TA, unit: meter; The distance calculated based on TA or path loss is the distance from the UE to the antenna port, which is a three-dimensional distance with an elevation angle. Normally, the UE's altitude is lower than the eNB's altitude, while the user's latitude and longitude change in two dimensions. The specific calculation of the distance from the UE to the base station is as follows: Ignoring the UE height, according to the Pythagorean theorem: L 2 +H 2 =d 2 The straight-line distance L from the UE to the antenna and the base station height H are derived from the engineering parameter base station height. ; The latitude and longitude information of the base station is obtained by obtaining the base station's parameters, and the latitude and longitude (X0, Y0) of the terminal user can be obtained by conversion. The specific three-point localization based on MR_TDOA is as follows: Extract the signal level of the primary serving cell and the signal level of at least two neighboring cells from the MR data. Using the wireless propagation model algorithm, obtain the distance d1 from the terminal to the primary serving cell and the distances d2 and d3 from the neighboring cell base stations, respectively. The location of the terminal user from the base station and neighboring base stations is obtained using the empirical formula for 4G path loss calculation. The empirical formula for 4G path loss calculation is as follows: L COST 231-Hata =46.3+33.9 log 10 (f c )-13.82 log 10 (h b )+(44.9-6.55 log 10 (h)) log 10 (d)+C M； Among them, f c The wireless signal frequency is 1500-2000MHz, in MHz; C M To cover the scene correction factor, cover_class covers rural areas, towns, general urban areas, and core urban areas, C M The values ​​are 0 / 3 / 6 respectively; h is the height difference between the terminal and the base station antenna, which is equal to the base station height in the engineering parameters by default, in meters; d is the distance between the base station antenna and the mobile station antenna, in kilometers. By using the least squares method combined with the base station's latitude and longitude information, the latitude and longitude (X1, Y1) of the terminal user can be obtained after conversion. The specific optimization of MRO localization results based on the fingerprint database is as follows: The MRO positioning result is mapped to the fingerprint database grid by minimizing the Euclidean distance algorithm. The grid whose feature information is closest to the feature information contained in the current MRO positioning result is found, and the position of this grid is finally used as the position of the corresponding terminal user. Simultaneously, the TA+AOA or TDOA three-point positioning results data under different coverage scenarios are continuously input into the OTT fingerprint database to build a machine learning model and calculate the MR positioning result offset parameters under different coverage scenarios.

2. A mobile network user positioning system based on the association of MRO and DPI data, characterized in that, This system is used to implement the mobile network user location method based on the association of MRO and DPI data as described in claim 1; the system includes, The fingerprint database establishment unit is used to obtain the user's latitude and longitude, i.e., OTT positioning, through user plane DPI data, establish a fingerprint database using the user's latitude and longitude extracted from OTT positioning, and use signaling plane DPI data to backfill user information from MRO data, and finally establish the association between MRO data and OTT fingerprint database. The positioning unit is used to locate the user's latitude and longitude using parameter information in MRO data through the TA+AoA or TDOA three-point positioning method; The tuning unit is used to effectively associate the MRO positioning results with the OTT fingerprint database, optimize and tune the MRO positioning results, and improve the accuracy of the MRO positioning results.

3. An electronic device, characterized in that, include: Memory and at least one processor; The memory contains computer programs; The at least one processor executes the computer program stored in the memory, causing the at least one processor to perform the mobile network user location method based on MRO and DPI data association as described in claim 1.

4. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that can be executed by a processor to implement the mobile network user location method based on MRO and DPI data association as described in claim 1.

Images