Fusion positioning method and system, electronic device, and storage medium

Abstract

Embodiments of the present application provide a fusion positioning method and system, electronic equipment and storage medium, belonging to the technical field of artificial intelligence. Based on multiple network positioning algorithms, the present application determines the positioning data stream of the corresponding algorithm under the rolling time window according to the obtained basic signaling data stream, then selects a positioning data stream as the target positioning data from the multiple positioning data streams according to the screening rule, determines the trajectory data of the mobile terminal according to the target positioning data stream, extracts the features of the trajectory data to obtain the multi-dimensional spatio-temporal trajectory features, and determines the fusion positioning information of the mobile terminal by taking the multi-dimensional spatio-temporal trajectory features as the input of the nodes of the Bayesian network model. By adopting multiple positioning algorithms and calculating the multi-dimensional spatio-temporal trajectory features after screening the positioning data, the fusion of multiple positioning methods is realized, and the position positioning accuracy is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of artificial intelligence, and particularly relates to a fusion positioning method and system, an electronic device and a storage medium. BACKGROUND

[0002] Location Based Services (LBS) means accurate positioning of a mobile terminal. The position information of the mobile terminal is obtained through positioning technology, and is provided to the mobile user himself or other communication systems, so as to realize various location-related services and provide diversified location services for users. With the rapid development of the Internet and the continuous improvement of communication technology, the technology for positioning the mobile terminal is changing rapidly. Under the background of the vigorous development of the Internet of Things, location-based services are rapidly developing and penetrating into all aspects of our lives, and the demand for high-precision positioning is increasing.

[0003] Although the operator network positioning technology and scale are developing rapidly at present, the positioning system still has a broad development space in practice. For example, a general positioning system has the disadvantages of high system complexity, high deployment cost, sensitivity to non-line-of-sight (NLOS) obstruction, weak robustness, etc. The positioning system of the operator at the present stage is roughly divided into three categories:

[0004] The first category is to realize accurate positioning by using assisted global positioning system (AGPS). The traditional GPS has the characteristics of wide coverage in outdoor environment, but the positioning accuracy is generally above one meter, which cannot meet the demand of high-precision positioning. At the same time, the satellite signal is easily affected by the sheltering objects such as buildings, trees and urban multipath environment, which leads to sharp decline of positioning accuracy or even cannot be positioned. In addition, the GPS system needs to search for positioning satellite ephemeris, so the hardware initialization time is long. The assisted global positioning system (AGPS) has the advantage of high positioning accuracy. Since the cellular network is introduced, the data transmission capability of the cellular network can make the positioning speed fast and the positioning accuracy higher. However, due to the influence of sheltering objects, some places cannot be positioned, which leads to small sample size.

[0005] The second type is to use the base station position to realize fast positioning, which can be divided into multi-base station and single-base station positioning according to the type. The multi-base station positioning technology is based on time of arrival (TOA) and time difference of arrival (TDOA), the number of base stations is 3 or more, and the positioning accuracy of the system can be improved by increasing the number of base stations, but the shape and position of the topology have higher requirements, and the synchronization between the base stations is also required for the TDOA-based positioning method. These reasons lead to high positioning cost of this type of method; the single-base station positioning technology is based on time of arrival (TOA) and angle of arrival (AOA), and the base station requires an antenna array to obtain the arrival phase information of the signal. This type of method greatly reduces the positioning cost without clock synchronization and meets the requirements of high-precision positioning, but the current single-base station positioning technology has two shortcomings: on the one hand, the existing single-base station technology is only applicable to line-of-sight (LOS) positioning scenarios, and in actual complex scenarios, signal non-line-of-sight (NLOS) propagation often occurs. Compared with the multi-base station system, the single tag in the single-base station system has only one signal link, which is more susceptible to NLOS interference, making it difficult to achieve robust positioning in actual complex scenarios; on the other hand, the positioning accuracy of the existing single-base station positioning technology in the coverage range of the single base station is basically fixed, and cannot increase the positioning accuracy in the service area by increasing the number of base stations like the existing multi-base station positioning technology, making it difficult to meet the demand of higher positioning accuracy scenarios.

[0006] The third type is the measurement report (MR) fingerprint positioning technology, which first collects each sample point in the positioning space for training, that is, collects the signal characteristics of each sample point to form a fingerprint database. If a certain to-be-positioned point is to be positioned, the signal characteristics of the point are collected and compared with the fingerprint database. The training sample point with the closest signal characteristics to the to-be-positioned point in the database is taken as the positioning result of the to-be-positioned point. However, the feature library fingerprint positioning method requires a very tedious data collection work, and since the fingerprint data usually accounts for a very small proportion, a sufficient number of data needs to be accumulated and updated over time and with network changes. In addition, due to the complex and variable radio propagation, the collection of location fingerprints itself is also not an easy problem.

[0007] Due to the data collection problem, the above three methods will have a certain degree of data loss, and pure use of a certain method cannot completely restore the position trajectory link of the user mobile terminal device (User Equipment, UE), resulting in low real-time positioning accuracy of the user mobile terminal device. SUMMARY

[0008] The main purpose of the embodiment of the present application is to propose a fusion positioning method, system, electronic device and storage medium, which aims to improve the accuracy of positioning of user mobile terminals.

[0009] To achieve the above object, a first aspect of the embodiment of the present application proposes a fusion positioning method, comprising the following steps:

[0010] Obtaining a basic signaling data stream of a mobile terminal;

[0011] Based on multiple network positioning algorithms, determining a positioning data stream of a corresponding algorithm under a rolling time window according to the basic signaling data stream;

[0012] According to a screening rule, selecting one positioning data stream from the multiple positioning data streams as a target positioning data stream;

[0013] Determining trajectory data of the mobile terminal according to the target positioning data stream;

[0014] Performing feature extraction on the trajectory data to obtain multi-dimensional spatio-temporal trajectory features;

[0015] Taking the multi-dimensional spatio-temporal trajectory features as the input of the nodes of a Bayesian network model to determine fusion positioning information of the mobile terminal.

[0016] In some embodiments, the fusion positioning method further comprises the following steps:

[0017] Based on a Flink streaming framework, writing different positioning data streams corresponding to different topics into Kafka queues;

[0018] Based on the Flink streaming framework, reading and processing the positioning data streams in the Kafka queues in parallel.

[0019] In some embodiments, the step of determining a positioning data stream of a corresponding algorithm under a rolling time window according to the basic signaling data stream based on multiple network positioning algorithms comprises the following steps:

[0020] Based on a base station positioning algorithm, determining a base station positioning data stream under a rolling time window according to signaling interface data streams and mobile management node data streams in the basic signaling data stream;

[0021] Based on a fingerprint positioning algorithm, determining a fingerprint positioning data stream under a rolling time window according to measurement report data streams in the basic signaling data stream.

[0022] In some embodiments, the step of selecting one positioning data stream from the multiple positioning data streams as a target positioning data stream according to a screening rule comprises the following steps:

[0023] For a current rolling time window, when there is AGPS positioning data, selecting a positioning data stream with the smallest deviation from the AGPS positioning data from the multiple positioning data streams as the target positioning data stream;

[0024] When the AGPS positioning data does not exist, an intermediate value is selected from the plurality of positioning data streams as the target positioning data stream.

[0025] In some embodiments, the determining trajectory data of the mobile terminal according to the target positioning data stream comprises the following steps:

[0026] The data in the target positioning data stream is preprocessed and integrated to obtain continuous signaling trajectories;

[0027] The distance between the start base station and the end base station in each signaling trajectory is calculated according to the continuous signaling trajectories;

[0028] The moving speed and acceleration are determined according to the distance of each signaling trajectory and the residence time of the mobile terminal in each signaling trajectory;

[0029] The moving direction is determined according to the continuous signaling trajectories;

[0030] The trajectory data is obtained according to the moving speed, the acceleration and the moving direction.

[0031] In some embodiments, the feature extraction of the trajectory data to obtain multi-dimensional spatio-temporal trajectory features comprises the following steps:

[0032] Based on a preset Gaussian membership function, the fuzzy statistical features are determined according to the moving speed and the acceleration, wherein the Gaussian membership function is established by prior knowledge and fuzzy theory, and is used to represent the mapping relationship between the moving speed and acceleration and the travel mode;

[0033] Based on the K-Means algorithm, the moving speed and the acceleration are respectively clustered and statistically analyzed to obtain clustering features;

[0034] The multi-dimensional spatio-temporal trajectory features are obtained according to the trajectory data, the fuzzy statistical features and the clustering features.

[0035] In some embodiments, the Bayesian network model is obtained by the following steps:

[0036] Based on the historical data of the mobile terminal, the multi-dimensional spatio-temporal trajectory features under different rolling time windows are determined, and the AGPS positioning data of the mobile terminal under the rolling time window is labeled to obtain training samples;

[0037] The multi-dimensional spatio-temporal trajectory features in the training samples are input into the Bayesian network model for forward transmission to obtain fusion positioning information;

[0038] Based on the same rolling time window, the Chebyshev distance is determined according to the corresponding fusion positioning information and the AGPS positioning data;

[0039] The Bayesian network model is evaluated and updated according to the Chebyshev distance.

[0040] To achieve the above object, a second aspect of the embodiment of the present application provides a fusion positioning system, comprising:

[0041] A first module is configured to acquire basic signaling data flow of a mobile terminal.

[0042] A second module is configured to determine positioning data flow of corresponding algorithm under a rolling time window according to the basic signaling data flow based on multiple network positioning algorithms.

[0043] A third module is configured to select one positioning data flow as target positioning data flow from the multiple positioning data flows according to a screening rule.

[0044] A fourth module is configured to determine trajectory data of the mobile terminal according to the target positioning data flow.

[0045] A fifth module is configured to perform feature extraction on the trajectory data to obtain multi-dimensional space-time trajectory features.

[0046] A sixth module is configured to determine fusion positioning information of the mobile terminal by taking the multi-dimensional space-time trajectory features as input of nodes of a Bayesian network model.

[0047] To achieve the above object, a third aspect of the embodiment of the present application provides an electronic device, comprising:

[0048] At least one processor;

[0049] At least one memory configured to store at least one program;

[0050] When the at least one program is executed by the at least one processor, the at least one processor is caused to implement the method of the first aspect.

[0051] To achieve the above object, a fourth aspect of the embodiment of the present application provides a storage medium, wherein a program executable by a processor is stored, and the program executable by the processor is used to implement the method of the first aspect when executed by the processor.

[0052] The technical scheme of the present application has at least one of the following advantages or beneficial effects: based on multiple network positioning algorithms, the positioning data stream of the corresponding algorithm in the rolling time window is determined according to the obtained basic signaling data stream, then a positioning data stream is selected from the multiple positioning data streams as the target positioning data according to the screening rule, the trajectory data of the mobile terminal is determined according to the target positioning data stream, the multi-dimensional space-time trajectory feature is obtained by feature extraction on the trajectory data, and the multi-dimensional space-time trajectory feature is used as the input of the node of the Bayesian network model to determine the fusion positioning information of the mobile terminal. By using multiple positioning algorithms and screening the positioning data to calculate the multi-dimensional space-time trajectory feature, the fusion of multiple positioning methods is realized, and the position positioning accuracy is improved. BRIEF DESCRIPTION OF DRAWINGS

[0053] Figure 1 is a fusion positioning method flowchart provided by an embodiment of the present application;

[0054] Figure 2 is a multi-dimensional space-time trajectory feature schematic diagram provided by an embodiment of the present application;

[0055] Figure 3 is a fusion positioning process schematic diagram under the Flink streaming framework provided by an embodiment of the present application;

[0056] Figure 4 is a terminal position trajectory updating process schematic diagram under a rolling time window provided by an embodiment of the present application;

[0057] Figure 5 is a fusion position model construction and updating process schematic diagram provided by an embodiment of the present application;

[0058] Figure 6 is a Bayesian network model training method flowchart provided by an embodiment of the present application;

[0059] Figure 7 is a fusion positioning system schematic diagram provided by an embodiment of the present application;

[0060] Figure 8 is an electronic device schematic diagram provided by an embodiment of the present application;

[0061] Figure 9 is a fusion positioning method provided by an embodiment of the present application Figure 1 is a flowchart of step S104 in the method;

[0062] Figure 10 is a flowchart of step S106 in the method. Figure 1 DETAILED DESCRIPTION

[0063] ​In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and not to limit the present application.

[0064] It should be noted that although the functional modules are divided in the device schematic diagram, and the logical sequence is shown in the flowchart, in some cases, the steps shown or described can be performed in a manner different from the module division in the device or the sequence in the flowchart. The terms "first", "second", etc. in the specification and claims and the above drawings are used to distinguish similar objects, and do not necessarily describe a specific order or sequence.

[0065] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application and are not intended to limit the present application.

[0066] First, the terms involved in the present application are analyzed:

[0067] The base station positioning algorithm includes the base station ID positioning algorithm and the distance-angle positioning algorithm. The base station ID positioning algorithm obtains the base station number and the cell number of the terminal reported by the wireless network, and the location service platform translates the cell number into latitude and longitude coordinates, thereby obtaining the position of the user mobile terminal. The distance-angle positioning includes common time of arrival (TOA), angle of arrival (AOA), and time difference of arrival (TDOA) and angle of departure (AOL) positioning techniques, and the core idea is to obtain the position of the user mobile terminal according to the straight-line distance and angle of the user mobile terminal to the access base station. The accuracy is relatively higher than that of the base station ID positioning algorithm.

[0068] Fingerprint positioning algorithm, input the signal strength of each cell received by the mobile terminal in the MR message into the feature library, wherein the fingerprint database stores discrete signal strength and position coordinates; match the received wireless signal strength with the fingerprint in the database, find the closest result for positioning.

[0069] Gaussian membership function, Gaussian membership function is an important concept in fuzzy set theory, which is a function used to describe attribute values. The definition and characteristics of Gaussian membership function make it a very important concept in fuzzy set theory, which can be used to represent the degree to which an object belongs to a fuzzy set, can be used to establish fuzzy models, can be used to identify fuzzy patterns, and can be used to solve fuzzy control problems.

[0070] The K-Means algorithm, i.e., a k-means clustering algorithm, is an iterative solution clustering algorithm. Clustering is a process of classifying and organizing data members in a data set that are similar in some aspects, and clustering is a technique of discovering such internal structures, and the clustering technique is often referred to as unsupervised learning. The k-means clustering is given a data point set and a required number of clusters k, which is specified by a user, and the k-means algorithm repeatedly divides the data into k clusters according to a distance function.

[0071] A Bayesian network, also known as a belief network or a directed acyclic graphical model, is a kind of probabilistic graphical model. A Bayesian network is a directed acyclic graph (DAG) composed of nodes representing variables and directed edges connecting the nodes. The nodes represent random variables, and the directed edges between the nodes represent the mutual correlation between the nodes (from the parent node to the child node), and the relationship strength is expressed by conditional probability, and the node without the parent node is expressed by prior probability. The node variable can be an abstraction of any problem. It is suitable for expressing and analyzing uncertain and probabilistic events, and can be applied to conditional decision-making depending on multiple control factors, and can make inferences from incomplete, inaccurate or uncertain knowledge or information.

[0072] Embodiments of the present application provide a fusion positioning method and system, electronic equipment and storage medium, aiming to improve the accuracy of positioning the user's mobile terminal.

[0073] The fusion positioning method provided by the embodiments of the present application can be applied to a terminal, can also be applied to a server side, and can also be software running in a terminal or a server side. In some embodiments, the terminal can be a smart phone, a tablet computer, a notebook computer, a desktop computer, etc. The server side can be configured as an independent physical server, can also be configured as a server cluster or a distributed system composed of multiple physical servers, can also be configured as a cloud server providing basic cloud computing services such as cloud service, cloud database, cloud computing, cloud function, cloud storage, network service, cloud communication, middleware service, domain name service, security service, CDN, and big data and artificial intelligence platform. In some embodiments, the fusion positioning method can be applied to an LBS network element on the network side of a communication system, thereby providing accurate terminal position information for a mobile user himself or other service systems.

[0074] Figure 1 is an optional flowchart of the fusion positioning method provided by the embodiments of the present application, Figure 1The method in the method can include but is not limited to including steps S101 to S106.

[0075] Step S101, acquiring the basic signaling data flow of the mobile terminal;

[0076] Step S102, determining the positioning data flow of the corresponding algorithm in the rolling time window according to the basic signaling data flow based on multiple network positioning algorithms;

[0077] Step S103, selecting one positioning data flow as the target positioning data flow from the multiple positioning data flows according to the screening rule;

[0078] Step S104, determining the trajectory data of the mobile terminal according to the target positioning data flow;

[0079] Step S105, performing feature extraction on the trajectory data to obtain multi-dimensional spatio-temporal trajectory features;

[0080] Step S106, determining the fusion positioning information of the mobile terminal by taking the multi-dimensional spatio-temporal trajectory features as the input of the nodes of the Bayesian network model.

[0081] In step S101 of some embodiments, the basic signaling data flow is a plurality of basic signaling data collected in real time. The basic signaling data is mainly derived from O-domain data, i.e., the data domain of operation support system, also known as operation domain, abbreviated as OSS. The O-domain has network data such as signaling, alarm, fault, network resource, etc. The collected basic signaling data includes user plane data and control plane data, specifically including signaling interface data, mobile management node data, measurement report data, and bill data, etc., the signaling interface data can be 5G control plane N1N2 data (i.e., 5G-N1N2 data), the mobile management node data is 4G control plane MME data (i.e., 4G-MME data), the measurement report data is 4G or 5G control plane MR data (i.e., at least one of 4G-MR or 5G-MR data), and the bill data is 4G-VoLTE data or 4G-CDR data in user plane, etc.

[0082] In step S102 of some embodiments, the network positioning algorithm can be a 4 / 5G base station positioning algorithm and a 4 / 5G fingerprint positioning algorithm, and the positioning data stream corresponding to the network positioning algorithm can be a 4G base station positioning data stream, a 5G base station positioning data stream, a 4G-MR fingerprint positioning data stream, and a 5G-MR fingerprint positioning data stream. The base station positioning algorithm can be a base station-based ToA positioning algorithm or a TDoA positioning algorithm, and the basic signaling data can provide the data basis for positioning. Specifically, the MR data required by the fingerprint positioning algorithm is derived from the UE-MR data of the LTE and NR control plane; the base station positioning data required by the base station positioning algorithm is derived from the signaling data in the 5G-N1N2 and 4G-MME. The present embodiment does not limit the way of obtaining the positioning data stream corresponding to each network positioning algorithm based on the basic signaling data stream. The rolling time window of the present embodiment can be 30 seconds or 60 seconds, and the present embodiment does not make specific limitations.

[0083] In step S103 of some embodiments, processing the basic signaling data stream based on different network positioning algorithms can obtain multiple positioning data streams corresponding to a mobile terminal. Since different network positioning algorithms have different advantages, multiple positioning information can comprehensively reflect the position of the mobile terminal. Based on the preset screening rule, the obtained multiple positioning data streams are screened to obtain the target positioning data stream of the mobile terminal. The screening rule of the present embodiment can be to take the intermediate value of multiple positioning data processed at the same time node as the target positioning data, and the positioning data stream obtained by processing the basic signaling data stream in the rolling time window using the network positioning algorithm corresponding to the target positioning data is the target positioning data stream. The screening rule can also be to take the positioning data stream closest to the AGPS positioning data as the target positioning data stream. The above screening rules are only exemplary, and the screening rule can also use other screening methods or combinations of different screening methods.

[0084] In step S104 of some embodiments, the trajectory data of the mobile terminal includes but is not limited to the moving speed, acceleration, moving direction, and distance to the access base station of the mobile terminal processed by the target positioning data stream in the rolling time window.

[0085] In steps S105 and S106 of some embodiments, referring to Figure 2 The multi-dimensional space-time trajectory feature includes a trajectory feature, a fuzzy statistical feature, and a clustering feature. The trajectory feature includes a moving speed, a moving direction, an acceleration, and a distance to an access base station. The fuzzy statistical feature includes a district importance and a travel mode. The clustering feature includes a center deviation of the moving speed or acceleration.

[0086] In some embodiments, the fusion positioning method of the embodiment of the present application further includes but is not limited to comprising steps S201 to S202:

[0087] In step S201, based on the Flink streaming framework, different positioning data streams are written into different Kafka queues of different topics;

[0088] In step S202, based on the Flink streaming framework, the positioning data streams in the Kafka queue are read in parallel and processed.

[0089] In this embodiment, the Flink streaming framework is used for the collection and data processing of the basic signaling data, which can improve the real-time performance of the location service network element for the mobile terminal location update. Referring to Figure 3 and Figure 4 The process of the fusion positioning method of the real-time example of the present application under the Flink streaming framework is as follows:

[0090] The multiple different positioning data streams are written into different Topics of Kafka, and the multiple positioning data includes 4G base station positioning data, 4G fingerprint positioning data, 5G base station positioning data and 5G fingerprint positioning data, etc.

[0091] Based on the rolling time window W under the real-time stream, each Topic in Kafka is read in parallel through the Flink Streaming, and the positioning data in the Topic is preprocessed by OneHot or binning, etc. The preprocessed positioning data is grouped according to the mobile terminal identifier, and the positioning data of the mobile terminal is sorted according to the timestamp. The accuracy level of the positioning data is marked according to the source, and each type of accuracy positioning data with the mobile terminal as the unique identifier is obtained.

[0092] Under the Flink streaming framework, the moving speed, moving direction, acceleration, and access sector distance of the mobile terminal are calculated according to the filtered target positioning data. The historical position, moving speed, moving direction, and acceleration of the same mobile terminal at the same time on weekdays or holidays are traversed. In addition, the fuzzy statistical features are obtained by applying the Gaussian membership function, and the clustering features are obtained by clustering the moving speed and acceleration using the spatial k-means. Considering the need to evaluate the Bayesian network model for feature fusion in the future, AGPS positioning data is collected synchronously and included in the cache. If there is AGPS positioning data in the corresponding rolling time window, it is marked as 1, otherwise it is marked as 0. It should be noted that the AGPS positioning data does not participate in the feature construction calculation. The moving distance, moving direction, moving speed, acceleration, fuzzy statistical features, and clustering features are used as multi-dimensional spatio-temporal trajectory features. The multi-dimensional spatio-temporal trajectory features are input into the trained Bayesian network model to predict the fusion positioning information of the mobile terminal. The model output result is written into the mobile terminal fusion positioning information real-time data stream Topic, and the position trajectory of the mobile terminal is updated based on the fusion positioning information.

[0093] The positioning data, trained Bayesian network model, and cache data relied on by the Flink streaming framework during execution are placed in the Elasticsearch cluster, and the model output result is written back to Elasticsearch for subsequent caching. The cache contains the latest information cache of all mobile terminals.

[0094] In some embodiments, step S102 can include but is not limited to steps S301 to S302:

[0095] Step S301, based on the base station positioning algorithm, determining the base station positioning data stream in the rolling time window according to the signaling interface data stream in the basic signaling data stream and the mobile management node data stream;

[0096] Step S302, based on the fingerprint positioning algorithm, determining the fingerprint positioning data stream in the rolling time window according to the measurement report data stream in the basic signaling data stream.

[0097] In this embodiment, since the user mobile terminal positioning data has a lagging characteristic, the collected positioning data at the current time is actually the real data at the last time. A plurality of network positioning algorithms are applied to optimize the current positioning data of the mobile terminal to obtain the best estimated positioning information corresponding to the last time, and the network positioning algorithm content is as follows:

[0098] Base station positioning algorithm, mobile network is connected by a base station, and a base station is divided into several cells. The operator collects 4G control plane MME and MR signaling data, 4G user plane VoLTE and CDR call data, 5G control plane N1N2 and MR signaling data, and the base station positioning algorithm is divided into base station ID positioning and distance angle positioning according to the number of base stations searched by the user terminal device and the accuracy requirement. The base station ID positioning obtains the base station number ENB_ID(4G) / GNB_ID(5G) and the cell number CELL_ID reported by the wireless network, and the location service platform translates the cell number into latitude and longitude coordinates to obtain the position of the user mobile terminal. The distance angle positioning includes common time of arrival (TOA), angle of arrival (AOA) and time difference of arrival (TDOA) and angle of departure (AOL) positioning technologies, and the core idea is to obtain the position of the user mobile terminal according to the straight line distance and angle geometry of the user mobile terminal to the access base station, and the accuracy is relatively higher than that of the base station ID positioning.

[0099] Fingerprint positioning algorithm, the signal strength of each cell received by the mobile terminal is input into the feature library, and the MR positioning data is calculated. Specifically, the signal strength of each cell received by the UE in the MR message (periodic and event type) is used as the input of the feature library, and the position of the UE is calculated. First, the field strength information is rasterized, that is, the network is divided into a plurality of square grids according to a certain scale, and the feature vector of each grid (5m×5m×5m) is simulated; then the MR cell field strength vector is constructed, and the vector distance between the MR belonging grid feature vector and the fingerprint library grid feature vector is used to determine the grid to which the MR belongs. The distance between the two vectors is calculated by using the Euclidean distance of the two vectors. The fingerprint features in the grid include the field strength distribution information of the adjacent cells, the mobility feature information, the number of adjacent cells, etc. In the positioning process, the fingerprint with the shortest vector distance between the MR point and the fingerprint in the fingerprint library is locked, and the grid position corresponding to the fingerprint can be considered as the positioning position of the corresponding MR.

[0100] In some embodiments, step S103 can include but is not limited to steps S401 to S402:

[0101] Step S401, for the current rolling time window, when there is AGPS positioning data, select the positioning data stream with the smallest deviation from the AGPS positioning data from the plurality of positioning data streams as the target positioning data stream;

[0102] Step S402, when there is no AGPS positioning data, select the intermediate value from the plurality of positioning data streams as the target positioning data stream.

[0103] In some embodiments, referring to Figure 9, step S104 can include but is not limited to including steps S501 to S505:

[0104] Step S501, pre-processing and integration of data in the target positioning data stream to obtain continuous signaling trajectories;

[0105] Step S502, calculating the distance between the start base station and the end base station in each signaling trajectory according to the continuous signaling trajectory;

[0106] Step S503, determining the moving speed and acceleration according to the distance of each signaling trajectory and the residence time of the mobile terminal in each signaling trajectory;

[0107] Step S504, determining the moving direction according to the continuous signaling trajectory;

[0108] Step S505, obtaining trajectory data according to the moving speed, acceleration and moving direction.

[0109] In this embodiment, the target positioning data of the mobile terminal under the rolling time window is pre-processed to eliminate invalid data, and continuous signaling trajectories are obtained. The distance between the start base station and the end base station in each signaling trajectory data is calculated, and the residence time of the mobile terminal in each signaling trajectory is calculated to determine the average moving speed and acceleration of the mobile terminal under the rolling time window. The moving direction is determined according to the continuous signaling trajectory. The moving speed, acceleration and moving direction are integrated to obtain the trajectory data.

[0110] In some embodiments, referring to Figure 10 , step S106 can include but is not limited to including steps S601 to S603:

[0111] Step S601, determining the fuzzy statistical features according to the moving speed and acceleration based on the preset Gaussian membership function, wherein the Gaussian membership function is established by prior knowledge and fuzzy theory, and is used to represent the mapping relationship between the moving speed and acceleration and the travel mode;

[0112] Step S602, respectively clustering and statistically analyzing the moving speed and acceleration based on the K-Means algorithm to obtain clustering features;

[0113] Step S603, obtaining multi-dimensional spatio-temporal trajectory features according to the trajectory data, the fuzzy statistical features and the clustering features.

[0114] In this embodiment, the Gaussian membership function corresponding to different travel modes and moving speed and acceleration is established based on the prior knowledge summary result and fuzzy theory, and the fuzzy statistical features of the terminal signaling are obtained through the Gaussian membership function. For example, the travel mode mapping relationship in the Gaussian membership function is shown in Table 1.

[0115] Table 1 travel mode mapping relationship

[0116]

[0117] The mobile speed and acceleration features are clustered by using a spatial K-Means algorithm to obtain clustering features of the mobile terminal, the obtained fuzzy statistical features and clustering features are added to the rolling time window mobile terminal trajectory data, and the distance, moving direction, average moving speed, acceleration, fuzzy statistical features and clustering features are used as multi-dimensional spatio-temporal trajectory features.

[0118] In some embodiments, with reference to Figure 6 , the Bayesian network model is obtained by the following steps:

[0119] In step S701, multi-dimensional spatio-temporal trajectory features under different rolling time windows are determined based on historical data of the mobile terminal, and AGPS positioning data of the mobile terminal under the rolling time window is labeled to obtain a training sample;

[0120] In step S702, the multi-dimensional spatio-temporal trajectory features in the training sample are input into the Bayesian network model for forward transmission to obtain fused positioning information;

[0121] In step S703, the corresponding fused positioning information and AGPS positioning data under the same rolling time window are determined to determine the Chebyshev distance;

[0122] In step S704, the Bayesian network model is evaluated and updated according to the Chebyshev distance.

[0123] In this embodiment, with reference to Figure 5 , by collecting the basic signaling data stream of the mobile terminal, the basic signaling data stream is processed in a streaming manner to obtain positioning data streams corresponding to different network algorithms, after feature extraction based on the positioning data streams, the Bayesian network model is trained offline in batches and online by using multi-dimensional spatio-temporal trajectory features.

[0124] A Bayesian network (Bayesian network) is also called a belief network (Belief Network), which is a graphical modeling method that uses a graph to represent a joint probability distribution function between a set of variables, and the network topology structure is a directed acyclic graph (DAG). Assuming that G=(I, E) represents a DAG, I is a set of all nodes in the graph, E is a set of all directed edges, and the function pa(x) represents a mapping from the child node x to the parent node. Let x i represent a random variable represented by a node i in the DAG, and for any x i , i∈I, the probability can be represented as:

[0125]

[0126] The Bayesian network model comprises a structure model and a set of conditional probability distribution functions corresponding to the structure model, and the strength of the relationship between variables in the network structure is determined by the conditional probability distribution functions expressed by the directed arcs. The fusion target point in the rolling time window is affected by multi-dimensional spatiotemporal trajectory features, the features are abstracted as nodes, and thus a fusion position model based on a Bayesian network is established.

[0127] For offline batch training of the Bayesian network model (i.e., the fusion position model), based on historical data, a rolling window containing AGPS markers is divided into a training set and a test set at a ratio of 7:3 or 8:2, multi-dimensional spatiotemporal trajectory features are abstracted as nodes, a directed arc is obtained based on a constrained Bayesian network structure learning algorithm, Bayesian network modeling is preliminarily completed on the training set, travel purposes and commuting characteristics are taken as deductive reasoning objects, and the Bayesian network probability model is perfected through a rule-based heuristic modeling method. In the model updating process, the fusion positioning information is corrected according to the time difference between the time stamp of the fusion positioning information and the time stamp of the AGPS positioning data, the average moving speed and the moving direction, the Chebyshev distance between the output of the corrected model and the AGPS positioning data is calculated, and the advantages and disadvantages of the fusion position model of the mobile terminal are evaluated.

[0128] For online training of the Bayesian network model, the fusion position model is online evaluated and updated based on the position prediction result output by the fusion position model and the current acquired position information at the previous time.

[0129] According to some specific embodiments of the present application, the construction process and application process of the fusion position model in the fusion positioning method of the embodiments of the present application are described:

[0130] S1, the 4 / 5G base station positioning algorithm and the fingerprint positioning algorithm are applied to obtain the original position information of the mobile terminal, i.e., the 4G base station positioning data, the 5G base station positioning data, the 4G MR positioning data and the 5G MR positioning data, the historical AGPS positioning data is synchronously collected, and is written into different Topics of Kafka;

[0131] S2, each Topic in Kafka is read in parallel through Flink Streaming, and the data in the Topic is preprocessed, and the position data accuracy level is marked according to the source;

[0132] S3, the Chebyshev distance deviation of each time stamp real AGPS positioning data and the four kinds of position data is compared, the data with the smallest deviation is selected as the unique value corresponding to the time stamp, and a real-time target positioning data stream is generated;

[0133] S4, the target positioning data stream is grouped according to a 30-second or 1-minute (i.e., a set period) rolling time window;

[0134] S5, sorting the target positioning data of the same mobile terminal appearing in the rolling time window according to the time stamp to form a position sequence of a single mobile terminal rolling time window;

[0135] S6, calculating the moving speed, moving direction, acceleration, access sector distance and other trajectory features of the mobile terminal in the rolling time window position sequence;

[0136] S7, applying a Gaussian membership function to obtain fuzzy statistical features, and using spatial k-means to cluster the moving speed and acceleration to obtain clustering features;

[0137] S8, dividing the rolling window containing the AGPS mark into a training set and a test set in a 7:3 (i.e. setting sample splitting) ratio;

[0138] S9, taking the trajectory features, fuzzy statistical features and clustering features as multi-dimensional spatio-temporal trajectory features, and abstracting them as nodes, and combining the historical AGPS positioning information to model on the training set through the Bayesian network algorithm;

[0139] S10, using Chebyshev distance to evaluate the pros and cons of the model on the test set, taking the travel purpose and commuting characteristics as the deductive reasoning object, and improving the Bayesian network model through the rule-based heuristic modeling method, that is, obtaining the trained fusion positioning model;

[0140] S11, using the fusion positioning model to output the mobile terminal fusion positioning information in the rolling time window, and outputting to various location service applications.

[0141] Reference Figure 7 , the embodiment of the application also provides a fusion positioning system, comprising:

[0142] The first module is used for acquiring the basic signaling data stream of the mobile terminal;

[0143] The second module is used for determining the positioning data stream of the corresponding algorithm under the rolling time window according to the basic signaling data stream based on multiple network positioning algorithms;

[0144] The third module is used for selecting one positioning data stream as a target positioning data stream from the multiple positioning data streams according to the screening rule;

[0145] The fourth module is used for determining the trajectory data of the mobile terminal according to the target positioning data stream;

[0146] The fifth module is used for extracting the multi-dimensional spatio-temporal trajectory features from the trajectory data;

[0147] The sixth module is used for taking the multi-dimensional spatio-temporal trajectory features as the input of the nodes of the Bayesian network model to determine the fusion positioning information of the mobile terminal.

[0148] It can be understood that the contents in the above fusion positioning method embodiments are applicable to the present system embodiment, the present system embodiment specifically implements the same functions as the above fusion positioning method embodiments, and achieves the same beneficial effects as the above fusion positioning method embodiments.

[0149] Please refer to Figure 8 , Figure 8 The hardware structure of the electronic device of another embodiment is illustrated, and the electronic device comprises:

[0150] The processor 801 can be implemented in the form of a general-purpose CPU (Central Processing Unit), a microprocessor, an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits, etc., for executing related programs to implement the technical solutions provided by the present embodiment;

[0151] The memory 802 can be implemented in the form of a ROM (ReadOnly Memory), a static storage device, a dynamic storage device, or a RAM (Random Access Memory), etc. The memory 802 can store an operating system and other application programs, and when the technical solutions provided by the present embodiment are implemented by software or firmware, the related program codes are saved in the memory 802 and called and executed by the processor 801 to implement the fusion positioning method of the present embodiment;

[0152] The input / output interface 803 is used to realize information input and output;

[0153] The communication interface 804 is used to realize the communication interaction between the present device and other devices, which can realize communication through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.);

[0154] The bus 805 transmits information between various components (such as the processor 801, the memory 802, the input / output interface 803, and the communication interface 804) of the device.

[0155] Among them, the processor 801, the memory 802, the input / output interface 803, and the communication interface 804 realize the communication connection between each other in the device through the bus 805.

[0156] The embodiment of the present application further provides a storage medium, which is a computer readable storage medium, used for computer readable storage, and stores one or more programs, which can be executed by one or more processors to realize the fusion positioning method.

[0157] The memory, as a non-transitory computer readable storage medium, can be used to store non-transitory software programs and non-transitory computer executable programs. In addition, the memory can include a high-speed random access memory, and can also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state memory device. In some embodiments, the memory can optionally include a memory remotely arranged relative to the processor, and these remote memories can be connected to the processor through a network. Examples of the network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0158] The embodiments described in the embodiments of the present application are used to more clearly illustrate the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. Those skilled in the art can know that, with the evolution of technology and the appearance of new application scenarios, the technical solutions provided by the embodiments of the present application are also applicable to similar technical problems.

[0159] Those skilled in the art can understand that, Figure 1 The technical solutions shown in the above description do not constitute a limitation on the embodiments of the present application, and can include more or fewer steps than those shown in the figures, or combine certain steps, or different steps.

[0160] The device embodiments described above are only schematic, and the units described as separate components can or can not be physically separate, that is, can be located in one place, or can be distributed on multiple network units. Part or all of the modules can be selected according to actual needs to achieve the purpose of the present embodiment.

[0161] Those skilled in the art can understand that all or some of the steps in the above disclosed method, the functions of the modules / units in the system and the device can be implemented as software, firmware, hardware, and appropriate combinations thereof.

[0162] The terms "first", "second", "third", "fourth", and the like in the description and in the claims of this application, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the use of the terms so termed is interchangeable under appropriate circumstances such that the embodiments of the application described herein are, for example, capable of orderly or chronological mundane operation, reverse order operation, based on circuitry availability, based on stated preference or the like, and that "default" or other orderings are thus permissible. Further, the terms "comprise", "comprising", "include", "including", and the like, are specifically intended to be open-ended. That is, references to individual steps and the like do not suhstantially exclude the presence of two or more of a given step or its integral presence in the process, method, system, article, or apparatus having been made with a wider scope. The use of notation such as "first", "second", "third", etc. does not generally limit the areas, but can be used for clarity, and merely establishes the order unless otherwise stated below.

[0163] It should be understood that, in the application, "at least one" means one or more, and "multiple" means two or more. "And / or" is used to describe the relationship between associated objects, which means that there can be three relationships, for example, "A and / or B" can mean that there are only A, only B, and A and B at the same time, where A and B can be singular or plural. The character " / " generally represents an "or" relationship between the associated objects. "At least one of the following" or the like means any combination of these items, including any combination of single or multiple items. For example, at least one of a, b or c can mean a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0164] In several embodiments provided in the application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the above units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be omitted or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed objects can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0165] The units described above as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, i.e. they can be located in one place or distributed on a plurality of network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0166] In addition, each of the functional units in the embodiments of the present application can be integrated in one processing unit, or each unit can exist alone physically, or two or more units can be integrated in one unit. The integrated unit can be implemented in the form of hardware, or in the form of a software functional unit.

[0167] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on such an understanding, the technical solutions of the present application, essentially or in part, or all or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes multiple instructions used to cause a computer device (such as a personal computer, a server, or a network device) to perform all or part of the steps of the methods in the embodiments of the present application. The foregoing storage medium includes: various types of U disks, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks or optical disks, and other storage media that can store programs.

[0168] The preferred embodiments of the embodiments of the present application are described above with reference to the accompanying drawings, and are not intended to limit the scope of the embodiments of the present application. Any modification, equivalent replacement and improvement made by those skilled in the art without departing from the scope and essence of the embodiments of the present application shall be within the scope of the embodiments of the present application.

Claims

1. A fusion positioning method, characterized by, The method comprises the following steps: obtaining basic signaling data flow of a mobile terminal; determining positioning data flow of corresponding algorithm under a rolling time window according to the basic signaling data flow based on multiple network positioning algorithms; selecting one positioning data flow as target positioning data flow from multiple positioning data flows according to a screening rule; the multiple positioning data flows comprise 4G base station positioning data flow, 4G fingerprint positioning data flow, 5G base station positioning data flow and 5G fingerprint positioning data flow; determining trajectory data of the mobile terminal according to the target positioning data flow; extracting features from the trajectory data to obtain multi-dimensional space-time trajectory features; the multi-dimensional space-time trajectory features comprise trajectory features, fuzzy statistical features and clustering features; the trajectory features comprise moving speed, moving direction, acceleration and access base station distance; the fuzzy statistical features comprise district importance and travel mode; and the clustering features comprise center deviation degree of moving speed or acceleration; determining fusion positioning information of the mobile terminal by taking the multi-dimensional space-time trajectory features as input of nodes of a Bayesian network model; the Bayesian network model is obtained by the following steps: determining multi-dimensional space-time trajectory features under different rolling time windows based on historical data of the mobile terminal and marking AGPS positioning data of the mobile terminal under the rolling time windows to obtain training samples; inputting the multi-dimensional space-time trajectory features in the training samples into the Bayesian network model for forward transmission to obtain the fusion positioning information; determining Chebyshev distance based on corresponding fusion positioning information and the AGPS positioning data under the same rolling time window; evaluating and updating the Bayesian network model according to the Chebyshev distance.

2. The fusion positioning method of claim 1, wherein, The fusion positioning method further comprises the following steps: based on a Flink streaming framework, corresponding different positioning data flows are written into Kafka queues of different topics; based on the Flink streaming framework, the positioning data flows in the Kafka queues are read in parallel and processed.

3. The fusion positioning method of claim 1, wherein, The method of determining positioning data flow of corresponding algorithm under a rolling time window according to the basic signaling data flow based on multiple network positioning algorithms comprises the following steps: determining base station positioning data flow under a rolling time window based on signaling interface data flow and mobile management node data flow in the basic signaling data flow based on a base station positioning algorithm; determining fingerprint positioning data flow under a rolling time window based on measurement report data flow in the basic signaling data flow based on a fingerprint positioning algorithm.

4. The fusion positioning method of claim 1, wherein, The method of selecting one positioning data flow as target positioning data flow from multiple positioning data flows according to a screening rule comprises the following steps: for a current rolling time window, when there is AGPS positioning data, selecting the positioning data flow with the minimum deviation from the AGPS positioning data as the target positioning data flow from the multiple positioning data flows; when there is no AGPS positioning data, selecting the intermediate value as the target positioning data flow from the multiple positioning data flows.

5. The method of claim 1, wherein, The method of determining trajectory data of the mobile terminal according to the target positioning data flow comprises the following steps: preprocessing and integrating data in the target positioning data flow to obtain continuous signaling trajectory; calculating a distance between a start base station and an end base station in each signaling trajectory according to the continuous signaling trajectories; determining a moving speed and an acceleration according to the distance of each signaling trajectory and a residence time of a mobile terminal in each signaling trajectory; determining a moving direction according to the continuous signaling trajectories; obtaining trajectory data according to the moving speed, the acceleration and the moving direction.

6. The fusion positioning method of claim 5, wherein, the feature extraction of the trajectory data to obtain multi-dimensional spatio-temporal trajectory features comprises the following steps: determining fuzzy statistical features according to the moving speed and the acceleration based on a preset Gaussian membership function, wherein the Gaussian membership function is established by prior knowledge and fuzzy theory, and is used to represent a mapping relationship between the moving speed and the acceleration and a travel mode; respectively clustering and statistically analyzing the moving speed and the acceleration based on a K-Means algorithm to obtain clustering features; obtaining multi-dimensional spatio-temporal trajectory features according to the trajectory data, the fuzzy statistical features and the clustering features.

7. A fusion positioning system characterized by, comprises: a first module configured to acquire basic signaling data streams of a mobile terminal; a second module configured to determine positioning data streams of corresponding algorithms in a rolling time window based on a plurality of network positioning algorithms according to the basic signaling data streams; a third module configured to select one positioning data stream as a target positioning data stream from a plurality of positioning data streams according to a screening rule, wherein the plurality of positioning data streams comprise 4G base station positioning data streams, 4G fingerprint positioning data streams, 5G base station positioning data streams and 5G fingerprint positioning data streams; a fourth module configured to determine trajectory data of the mobile terminal according to the target positioning data stream; a fifth module configured to extract features of the trajectory data to obtain multi-dimensional spatio-temporal trajectory features; the multi-dimensional spatio-temporal trajectory features comprise trajectory features, fuzzy statistical features and clustering features, the trajectory features comprise a moving speed, a moving direction, an acceleration and an access base station distance, the fuzzy statistical features comprise a district importance degree and a travel mode, and the clustering features comprise a center deviation degree of the moving speed or the acceleration; a sixth module configured to determine fusion positioning information of the mobile terminal by taking the multi-dimensional spatio-temporal trajectory features as inputs of nodes of a Bayesian network model; the Bayesian network model is obtained by the following steps: determining multi-dimensional spatio-temporal trajectory features in different rolling time windows and labeling AGPS positioning data of the mobile terminal in the rolling time windows based on historical data of the mobile terminal to obtain training samples; inputting the multi-dimensional spatio-temporal trajectory features in the training samples into the Bayesian network model for forward transmission to obtain the fusion positioning information; determining a Chebyshev distance based on corresponding fusion positioning information and the AGPS positioning data in the same rolling time window; evaluating and updating the Bayesian network model according to the Chebyshev distance.

8. An electronic device, comprising: comprises: at least one processor; at least one memory configured to store at least one program; when the at least one program is executed by the at least one processor, the at least one processor implements the fusion positioning method according to any one of claims 1 to 6.

9. A computer readable storage medium having stored therein a program which is executable by a processor, characterized in that, The program executable by the processor, when executed by the processor, is used to implement the fusion positioning method according to any one of claims 1 to 6. The processor can execute the program, and when the program is executed by the processor, the program is used to implement the fusion positioning method according to any one of claims 1 to 6.

Images