Electronic map processing method, electronic device and computer program product

Abstract

The invention provides an electronic map processing method, a communication device, electronic equipment and a computer program product, belongs to the technical field of electronic maps, and aims to solve the problems of long time consumption, high cost and the like caused by untimely map updating, inaccurate data and dependence on special equipment in the prior art. The method comprises the following steps: acquiring positioning data of a terminal in a communication network; and determining behavior mode characteristics of the terminal based on the positioning data, obtaining target geographic element information through the behavior mode characteristics, and updating the electronic map according to the target geographic element information.

Description

Technical Field

[0001] This application relates to the field of electronic map technology, and in particular to an electronic map processing method, communication device, electronic device and computer program product. Background Technology

[0002] Communication network nodes integrating sensing and communication functions, such as integrated sensing base stations, may misidentify non-interested targets as targets during target detection due to the inherent physical characteristics of their antenna beams, resulting in numerous erroneous identification trajectories. To distinguish between real and interfering targets, semantically informative environmental maps are introduced, allowing for filtering of identification results based on the marked geographic feature attributes. However, the accuracy of these environmental maps heavily relies on external map data, multi-camera aerial photographs, and the node's own engineering parameters. These data often suffer from update delays, incomplete coverage, and insufficient precision, leading to a mismatch between the map and the actual environment, thus weakening the reliability of target selection. Summary of the Invention

[0003] This application provides a method, system, electronic device, and vehicle for processing electronic maps, which can solve the problems of untimely map updates, inaccurate data, and long processing time and high cost due to reliance on dedicated equipment.

[0004] In a first aspect, embodiments of this application provide a method for processing electronic maps, the method comprising the following steps: Obtain the location data of the terminal in the communication network; Based on the location data, the behavioral pattern characteristics of the terminal are determined, target geographic element information is obtained through the behavioral pattern characteristics, and the electronic map is updated according to the target geographic element information.

[0005] Secondly, embodiments of this application provide a communication device, including the following: The communication unit is configured to wirelessly communicate with one or more terminals in a communication network and acquire the location data of the terminals; The sensing unit is configured to transmit sensing signals and receive their echo signals to acquire target trajectory data within the monitoring space; and The processing unit is connected to the communication unit and the sensing unit and is configured to update the electronic map according to the processing method described in the first aspect above, and to process the target trajectory data using the updated electronic map.

[0006] Thirdly, embodiments of this application provide an electronic device including a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the method described in the first aspect.

[0007] Fourthly, embodiments of this application provide a computer program product, the computer program product including a computer program stored on a non-transitory computer-readable storage medium, the computer program including program instructions, which, when executed by a computer, implement the steps of the method described in the first aspect.

[0008] In this embodiment, by acquiring the positioning data of the terminal in the communication network, target geographic element information is obtained based on the positioning data, and the electronic map is updated according to the target geographic element information. This achieves the dynamic self-updating capability of the electronic map, eliminating the need for a passive mode of periodic manual surveying or satellite revisiting. Through continuous analysis of terminal behavior data, the electronic map possesses the ability to self-evolve and update in real time, helping to solve the difficulties of delayed and incomplete electronic map updates caused by the rapid pace of urban construction. By analyzing the real behavior patterns of the terminal (such as vehicle trajectory and crowd gathering) to inversely derive the functional semantics of geographic elements, the system overcomes the problems of misjudgment and omission caused by occlusion, shadows, and viewing angles when relying solely on satellite imagery for image recognition. This achieves data-driven geographic element discovery, improving the semantic richness and accuracy of the electronic map. Utilizing the communication terminal as a mobile sensor eliminates the need for expensive professional surveying equipment and manual exploration costs, reducing the professional requirements for map updates and solving the problems of high costs and limited large-scale promotion of existing technologies. This significantly reduces the update cost of electronic maps and greatly improves their update efficiency. The location data of the terminal exists naturally with the coverage of the communication network, which allows the technology to be expanded to the entire network at low cost and adapt to various scenarios from cities to rural areas, overcoming the limitations of existing technologies that can only deal with single-point problems and cannot be deployed across the entire network. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a flowchart illustrating an electronic map processing method provided in an embodiment of this application; Figure 2This is a flowchart illustrating another electronic map processing method provided in an embodiment of this application; Figure 3 This is a flowchart illustrating another electronic map processing method provided in an embodiment of this application; Figure 4 This is a schematic diagram of the structure of a communication device provided in an embodiment of this application; Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0011] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0012] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0013] Harmonized Communication and Sensing (HCS) is one of the main new application scenarios added in 5G-Advanced (5G-A). 5G-A integrates the capabilities of traditional radar into communication base stations, achieving both communication and sensing functions through a single device (i.e., the integrated communication and sensing base station). The communication cell performs communication functions through communication signals, while the sensing cell performs sensing functions by transmitting sensing signals and receiving echo signals (i.e., the reflected signals generated after the sensing signals reach the target). Specifically, distance R can be distinguished by different echo times, different angles A can be distinguished by beam scanning, and different velocities V can be distinguished by Doppler, ultimately achieving drone sensing and positioning. The 5G-A integrated communication and sensing sensing cell has a mechanical tilt angle of 0 degrees, a total of 4 sensing beams, and covers the airspace within 300 meters above the base station, with a horizontal field of view (FOV) of 110 degrees and a vertical field of view of 36 degrees. In reality, even sensing cells with integrated sensing capabilities exhibit numerous ground lobes, which can misidentify moving objects such as vehicles and pedestrians as targets. This leads to many false alarm trajectories, known as Grating-lobe Target Recognition (GTR), where non-drone trajectories are mistakenly reported as drone trajectories. To address this specific scenario, a 3D environmental semantic map is introduced at the integrated sensing base station. Based on this map, the base station can identify which areas constitute the boundaries of roads, residential areas, and other properties. False alarms generated by vehicles, pedestrians, and other targets within these areas can be proactively suppressed and not reported.

[0014] Current 3D environmental semantic maps heavily rely on 5m satellite electronic maps and high-precision engineering parameters reported by antenna information sensor units (AISUs). However, with the rapid pace of urban construction, large vacant lots are being developed into roads, residential areas, office buildings, etc. Inaccurate, outdated, or unmapped 5m satellite maps provided by map providers significantly reduce the effectiveness of 3D environmental semantic maps in suppressing false alarms. To improve the effectiveness of 3D environmental semantic maps in suppressing false alarms, two methods are generally adopted: First, for areas with concentrated false alarm trajectories generated by UAV testing, combined with on-site surveys, if a specific area is confirmed to be a newly constructed road, industrial park, etc., the 3D environmental semantic map is manually reconstructed, and the sensing trajectories within that GTR area are removed. The drawback of this method is that it is only suitable for handling single-point problems and cannot be widely applied across the entire network. Second, surveying-grade UAVs are used to conduct aerial photography within a 2km radius of the integrated sensor construction area. A new electronic map is regenerated based on the aerial photographs with latitude and longitude information, and then the 3D environmental semantic map is reconstructed. This method relies on professional surveying UAVs and pilots, and is time-consuming and costly.

[0015] Current sensing base stations introduce 3D environmental semantic maps to address the false alarm problem in grating lobe target recognition. However, the inaccuracy of 5m satellite maps significantly reduces the false alarm suppression effect. Existing methods, such as manual removal of false alarm trajectories and reconstruction of 5m satellite images from UAV aerial photography, have limitations in terms of economy, timeliness, and complexity. The method of manually removing false alarm trajectories based on clustered areas requires determining the clustered areas through backend false alarm trajectory location data and confirming this through on-site surveys. For example, if the trajectories are distributed along newly constructed roads or riverside paths, or concentrated within a residential area, then GTR regions are manually drawn in the 3D semantic map, and sensing trajectories generated in these areas are treated as false alarms and not reported. This method requires a combination of frontend flight testing, backend data analysis, secondary on-site survey confirmation, and manual drawing of the 3D environmental semantic map, demanding strong professional capabilities and making large-scale deployment impossible. Reconstructing a 3D environmental semantic map from aerial photographs taken by surveying-grade drones, taking a 20 square kilometer area as an example, requires one drone, one professional pilot, and one test vehicle (including the driver). It involves taking 200m x 200m images in both horizontal and vertical directions, and is estimated to take approximately 5 days, generating a total of 300GB of aerial photographs. Simultaneously, a 3D raster map is generated based on the aerial photographs, ultimately producing a 3D environmental semantic map. This method requires acquiring a professional drone and pilot, is time-consuming and costly, and requires strong technical expertise to construct a 3D raster map from aerial photographs, presenting a high barrier to entry.

[0016] Therefore, there is an urgent need for an electronic map processing method to solve the problems of untimely map updates, inaccurate data, and high time consumption and cost due to reliance on special equipment in existing methods.

[0017] The following is in conjunction with the appendix Figures 1 to 5 This application provides a detailed description of an electronic map processing method, system, electronic device, and computer program product through specific embodiments and application scenarios.

[0018] Figure 1 This application illustrates an embodiment of an electronic map processing method, which can be executed by an electronic device, including a server and / or terminal devices. In other words, the method can be executed by software or hardware installed on the server and / or terminal devices, and includes the following steps: Step 110: Obtain the location data of the terminal in the communication network, the location data being used to provide the geographical location information of the terminal.

[0019] The term "terminal" refers to a mobile terminal, including but not limited to in-vehicle terminals and personal terminals. In-vehicle terminals are those installed in vehicles, such as 4G / 5G new energy vehicle infotainment systems. Personal terminals are terminals carried and used by individuals, such as smartphones, tablets, and portable hotspots.

[0020] Location data refers to latitude and longitude coordinates obtained through positioning technologies (such as GPS and base station positioning), primarily used for navigation and location services. For example, it includes real-time coordinate data calculated by mobile terminals like smartphones using GPS modules or base station signals. Location data encompasses, but is not limited to, Measurement Report (MR) data, GNSS (GPS / BeiDou) data, Wi-Fi positioning data, Bluetooth beacon data, and any data that provides geographic location information. MR data, in particular, is a dataset used for network evaluation and optimization in mobile communication systems. It is collaboratively generated by Mobile Stations (MS) and Base Transceiver Stations (BTS) and uploaded at fixed time intervals. The data typically undergoes basic processing at the Base Station Controller (BSC) to provide a basis for algorithms such as handover decisions and power control. By collecting real-time measurement data from all users during calls, MR data can replace some routine testing, reducing operational costs. It can analyze wireless coverage quality, user distribution, and behavioral patterns, and can be applied to scenarios such as handover process optimization and malicious network behavior investigation.

[0021] Step 130: Determine the behavioral pattern characteristics of the terminal based on the positioning data, obtain target geographic element information through the behavioral pattern characteristics, and update the electronic map according to the target geographic element information.

[0022] The behavioral pattern features include at least one of kinematic features and spatiotemporal clustering features. These features are not single indicators, but rather a multi-dimensional set of features derived from the historical location data of a group of terminals, revealing their movement patterns and spatiotemporal clustering patterns. For example, the kinematic features of a vehicle-mounted terminal can be used to complete the features of a road; these features describe the regularity of movement and are used to determine whether the movement trajectory conforms to the vehicle's driving pattern on the road. The spatiotemporal clustering features of a resident user's personal terminal can be used to complete the features of the Area of ​​Interest (AOI); these features describe clustering in the spatiotemporal dimension and are used to discover the regularity of people staying at specific times and locations.

[0023] The information on target geographic elements obtained through behavioral pattern features can be derived based on specific algorithmic models. Specifically, kinematic features (continuous, directional, and smooth trajectories) can be used as digital representations of linear geographic elements (i.e., roads). Algorithms such as clustering and boundary generation can be used to restore these abstract "motion lines" to concrete "road lines" on the map. Spatiotemporal clustering features (dense distribution of users within a specific spatiotemporal range) can be used as digital representations of area geographic elements (e.g., interest areas). Density clustering and boundary fitting can be used to restore these scattered "distribution points" to meaningful "regional areas" on the map.

[0024] The geographic feature information includes geometric information and attribute information. Geometric information includes spatial location and shape, such as the centerline coordinates and boundary coordinates (polygon border) of a new road, and the outline boundary coordinates of a new property. Attribute information includes semantic type and topological relationships. The semantic type is used to infer whether the geographic feature is a road, residential area, industrial park, commercial center, etc., while the topological relationships are used to determine how the newly discovered road connects to the existing road network and which roads surround the newly discovered property.

[0025] This application's embodiments, through continuous analysis of terminal behavior data, enable electronic maps to possess self-evolving and real-time updating capabilities. This breaks away from the passive mode of traditional map updates, which relies on periodic manual surveying or satellite revisits. It directly solves the problems of delayed updates and incomplete coverage caused by rapid urban development, improving the semantic richness and accuracy of electronic maps. The dynamic self-updating capability of electronic maps no longer simply identifies graphics from images, but rather infers the functional semantics of geographic elements (e.g., distinguishing roads and residential areas) by analyzing real-world terminal behavior patterns (such as vehicle trajectories and crowd gatherings). This overcomes the misjudgments and omissions caused by occlusion, shadows, and viewing angles when relying solely on satellite imagery for image recognition. This application's embodiments can discover "hidden" geographic elements not yet recorded by traditional surveying methods, such as newly constructed but unmapped roads, newly formed commercial or residential areas, filling map gaps in suburban and rural areas where traditional surveying coverage is insufficient, and achieving data-driven geographic element discovery. By utilizing existing communication terminals as "mobile sensors," the cost of expensive specialized surveying equipment (such as surveying drones) and manual surveying is eliminated. This reduces the professional requirements for map updates and overcomes the economic bottleneck of existing technologies (such as drone aerial photography) being too costly and unable to be widely deployed, significantly lowering the cost of updating electronic maps. The terminal's location data is naturally present along with the communication network coverage, allowing the technology to be expanded to the entire network at low cost and adaptable to various scenarios from urban to rural areas. This overcomes the limitations of existing solutions (such as manual surveying) that can only handle single-point problems and cannot be deployed across the entire network, demonstrating high scalability and adaptability.

[0026] Figure 2 This diagram illustrates a flowchart of another electronic map processing method provided by an embodiment of this application. This method can be executed by an electronic device, which may include a server and / or terminal devices. In other words, the method can be executed by software or hardware installed on the server and / or terminal devices, and includes the following steps: Step 210: Obtain the location data of the terminal in the communication network, wherein the location data is used to provide the geographical location information of the terminal.

[0027] Step 210 can be found above. Figure 1 The specific description of step 110 in the illustrated embodiment can achieve the same technical effect, and will not be repeated here to avoid repetition.

[0028] Step 220: Based on step 120 of the above embodiment, the method of determining the behavioral pattern characteristics of the terminal based on the positioning data, obtaining target geographic element information through the behavioral pattern characteristics, and updating the electronic map according to the target geographic element information may further include the following specific steps: Trajectory data from the vehicle terminal is selected from the positioning data. Based on the kinematic features of the trajectory data, road boundary information missing from the electronic map is generated. The road boundary information is then used as the target geographic feature information and topologically connected to the existing road network in the electronic map.

[0029] The trajectory data of the vehicle terminal can identify the measurement report data of the vehicle user based on the positioning data. The vehicle user can be identified according to the model approval number in the International Mobile Equipment Identity of the terminal, and potential vehicle users can be identified and added to the vehicle user database based on the terminal location change being less than a threshold within a preset time period.

[0030] Taking 4G / 5G new energy vehicle in-vehicle infotainment system users as an example, the IMEI-TAC (Type Allocation Code) is used to identify these users in existing communication networks. The IMEI is a unique digital code used to identify the mobile phone / terminal, while the TAC is used to uniquely identify the device model. It should be noted that any in-vehicle terminal with an IMEI number that can generate MR (Location Mapping) location data is acceptable; it does not necessarily have to be a new energy vehicle.

[0031] Ensure that the user's IMEI-TAC reporting switch is enabled on both the core network and base station sides of the communication network. The IMEI encoding typically includes the following:

[0032] The communication network extracts MR data from more than a week ago. The MR data contains information such as timestamps, IMEI-TAC, and latitude and longitude (generally implemented based on triangulation algorithms).

[0033] Derive the TAC prefixes of new energy vehicle manufacturers (e.g., Tesla: 86508305, BYD: 86708704) from the ITU IMEI database (e.g., GSMA IMEI Allocation Database).

[0034] Use the following regular expression to match IMEI-TAC:

[0035] By combining underground parking garage information from electronic maps (e.g., satellite maps), users who remain stationary for extended periods between 0:00 and 6:00 AM (this timeframe is adjustable and not mandatory) are added to the IMEI-TAC database if their IMEI-TAC is not already in the database. This ensures the real-time nature and completeness of the data. The specific method is as follows: (1) Judgment of stationary vehicle-mounted users: For the MR location data of all users from 0:00 to 6:00 AM, if the following conditions are met, they are marked as potential new energy vehicle-mounted users:

[0036] in, The timestamp difference (e.g., 4 hours). For changes in latitude and longitude distance (e.g., 10 meters). and Adjustments can be made as needed.

[0037] (2) Supplement the IMEI-TAC database: If the IMEI-TAC of the potential vehicle infotainment user is not in the database... TAC 新能源车 If it is in the library, then add it to the TAC library.

[0038] The kinematic features can be calculated based on trajectory data through kinematic feature modeling. These kinematic features include at least one of velocity, acceleration, rate of change of direction, and trajectory curvature. Combining existing road information from the electronic map (road direction, width, traffic light information, mainly used for acceleration calculation, interest surface bounding, road enclosure information, etc.) with the MR location information of the new energy vehicle's infotainment system (timestamp, latitude and longitude, etc.), kinematic feature modeling of the vehicle's user is performed, including direction, acceleration, velocity, trajectory smoothness, etc. Simultaneously, a Long Short-Term Memory (LSTM) network algorithm is used for enhancement processing. The kinematic feature calculation can be divided into data preprocessing and vehicle user motion feature extraction, with the specific steps as follows: Data preprocessing (1) Input: MR location information data of new energy vehicle (time stamp, latitude and longitude).

[0039] (2) Output: Trajectory point sequence , A timestamp at a specific moment. , To match the latitude and longitude at that moment, an LSTM-based trajectory prediction model is introduced to interpolate and compensate for missing or abnormal location points, thereby improving the integrity of the trajectory.

[0040] Vehicle-mounted infotainment system user motion feature extraction The system is characterized primarily by four dimensions: vehicle speed, acceleration, rate of change of direction, and trajectory curvature. The range is derived based on communication network data. Details are as follows: (1) Speed ​​calculation, refer to the following formula: And derive its range of values. (2) Acceleration calculation, refer to the following formula: And determine its value range. Generally, at intersections equipped with traffic lights, the acceleration drops to 0. (3) Rate of change of direction, refer to the following formula: And derive its range of values, where The trajectory direction angle (calculated using latitude and longitude). By capturing the direction sequence patterns of long-distance trajectories, abnormal direction abrupt changes (such as non-road turning behavior) are identified and eliminated.

[0041] (4) Trajectory curvature. Trajectory curvature reflects the degree of sharpness of the turn. The smaller the change in curvature, the smoother the trajectory. Refer to the following formula: And derive its range of values, where, For the velocity component (first derivative). For acceleration components (second derivative).

[0042] The road boundary information is used to complete the missing road information in the electronic map.

[0043] In this embodiment, road information is obtained by analyzing the actual movement trajectories of real vehicles. Since vehicle trajectories are naturally confined to the road area, and their motion characteristics (such as speed and acceleration) are strongly correlated with road structures (such as intersections and curves), the generated road boundary information has extremely high accuracy and authenticity, enabling high-precision and automated completion of the road network. This method can discover newly opened roads that have not yet been recorded by satellite or aerial maps, and even unnamed temporary roads or rural paths. As long as vehicles are traveling on them, they can be captured by the system and added to the map, solving the data loss problem caused by map update delays in suburban areas and newly built urban areas. It ensures the network topology connectivity of the completed roads and gives electronic maps the ability to discover "hidden roads." Through topological connections, it ensures that newly generated road segments are not isolated but can be organically integrated with the existing road network to form a complete and usable road network, ensuring the practicality and consistency of the updated map. Utilizing widely available vehicle-mounted terminals as mobile mapping units eliminates the need for professional mapping fleets or drone aerial photography, significantly reducing the cost of collecting high-precision road network data. Furthermore, the entire process from the opening of a new road to its use by vehicles, and then to its recognition by the system and updating to the electronic map, can be very short (e.g., a few days or hours), achieving near real-time road information updates.

[0044] In yet another exemplary embodiment, based on step 220 of the above embodiment, the method of selecting trajectory data from the vehicle terminal from the positioning data and generating the missing road boundary information of the electronic map based on the kinematic features of the trajectory data may further include the following specific steps: Calculate the kinematic features of the trajectory data; cluster the trajectory points in the trajectory data based on the density clustering algorithm and the calculated kinematic features, and identify the missing candidate road areas in the electronic map based on the clustering results; perform directional consistency screening on the identified candidate road areas to determine a continuous first road segment; generate the road boundary information based on the trajectory point set of the first road segment using the convex hull algorithm.

[0045] The trajectory points of the vehicle-mounted terminal can be obtained based on the historically accumulated MR location information (timestamp, latitude and longitude) generated by the vehicle-mounted terminal user. If the road information corresponding to the trajectory points is missing in the electronic map, new road information is generated by combining the vehicle-mounted terminal user's kinematic characteristics (vehicle speed, acceleration, rate of change of direction, trajectory curvature, etc.). This mainly involves adding road location borders, such as intersections equipped with traffic lights. A density-based spatial clustering of applications with noise (DBSCAN) algorithm is used to cluster the trajectory points, and a convex hull algorithm is used to generate the borders, with reasonable buffers set to make them more consistent with the actual road borders. The specific steps are as follows: Trajectory data preprocessing (1) Outlier filtering: Remove points whose speed or acceleration exceeds the physical limit (e.g., instantaneous speed > 150 km / h).

[0046] (2) Time window sliding: Smooth the continuous trajectory points according to the time window (e.g., 5 minutes) to eliminate noise from position information jitter.

[0047] Road boundary inference (1) Trajectory point clustering: DBSCAN is used to cluster trajectory points to identify high-frequency traffic areas missing in the electronic map.

[0048]

[0049] in, The neighborhood radius, min_samples The minimum number of cluster points can be adjusted.

[0050] (2) Orientation consistency screening: For each cluster, calculate the variance of the orientation angle. :

[0051] like If the threshold is less than the threshold (which can be adjusted as needed), the road segment is determined to be continuous, and the boundary point set is extracted. .

[0052] (3) Generation of road borders Convex Hull Algorithm: For the identified potential road region trajectory point set Calculate the convex hull (ConvexHull(B)) and generate the polygonal bounding box.

[0053] Buffer zone expansion: Based on empirical values ​​for road width (e.g., a typical width of 20 meters for a two-way four-lane road), expand the buffer zone outwards:

[0054] in Let be the midline point of the trajectory. Width This represents the total width of the road.

[0055] In this embodiment, density clustering algorithm can efficiently identify potential road areas with dense vehicle trajectories. Then, direction consistency filtering is used to eliminate random and chaotic trajectory interference, accurately extracting continuous road segments. Finally, convex hull algorithm is used to automatically generate road boundaries that fit the actual situation. This achieves fully automatic and high-precision conversion from raw trajectory data to standardized road geometry information, significantly improving the efficiency and accuracy of road information completion in missing areas of electronic maps. At the same time, it effectively overcomes the dependence of traditional methods on manual surveying or high-precision imagery.

[0056] In yet another exemplary embodiment, based on step 220 of the above embodiment, the method of connecting the road boundary information with the existing road network in the electronic map may further include the following specific steps: generating a new road segment based on the road boundary information, connecting the new road segment with the existing road network in the electronic map, and verifying the connection.

[0057] The topology connection is used to connect newly identified road segments with the existing electronic map road network topology to ensure connectivity.

[0058] The verification process may include trajectory smoothness verification and satellite map road trajectory alignment.

[0059] Trajectory smoothness verification includes calculating the trajectory smoothness score of the newly added road segment (refer to the trajectory curvature calculation formula in step 220). If the curvature of the newly added road trajectory... If the curvature is less than the threshold (adjustable), it is considered a valid road; if the curvature of the newly added road trajectory is... If the value exceeds the threshold, it is determined to be an invalid road (abnormal data) and deleted directly.

[0060] Satellite map road trajectory alignment is used to overlay the generated road borders with the satellite map, and the geometric consistency is manually verified.

[0061] By combining existing road network information from electronic maps with newly added road location information, a road network dataset is formed. It includes road name, coordinates (set of line segments), and topological relationship.

[0062] In this embodiment, new roads are seamlessly integrated with the existing road network through topological connections, ensuring the integrity and availability of road network data. Then, through a dual verification mechanism of trajectory smoothness verification and satellite map alignment, invalid roads caused by data anomalies are effectively screened out, ensuring the geometric rationality and consistency with reality of the completed roads. Finally, a high-quality road network dataset containing complete topological relationships is generated, which greatly improves the accuracy and reliability of electronic maps.

[0063] In yet another exemplary embodiment, based on step 120 of the above embodiment, the method of determining the behavioral pattern characteristics of the terminal based on the positioning data, obtaining target geographic element information through the behavioral pattern characteristics, and updating the electronic map according to the target geographic element information may further include the following specific steps: Based on the time information of the location data, the location data of the target user in the communication network is divided into datasets for different time periods, and the datasets for each time period are clustered to determine the target user's permanent residence area in different time periods; based on the permanent residence area and the road information of the electronic map, the missing interest surface information of the electronic map is generated, and the interest surface information is used as the target geographic feature information.

[0064] The time information can be obtained based on the timestamp of the MR data. The timestamp is mainly used to identify the exact time point when the data was collected or the event occurred.

[0065] The target users can include resident users in the communication network. The corresponding time and location data can be obtained through the MR data provided by the resident users' personal terminals.

[0066] Areas of Interest (AOIs), also known as information surfaces, are data units in electronic maps that describe regional geographic entities. They typically include basic information such as name, address, category, and latitude and longitude coordinates. As a supplement to Points of Interest (POIs), AOIs express geographic spatial extent more accurately through areal data. They are suitable for regional entities such as residential areas, universities, and industrial parks, and offer higher stability, reflecting lower-frequency geographic changes. POIs, on the other hand, abstract higher-level geographic elements using point data. Together, they constitute the basic representation of map data.

[0067] The spatiotemporal clustering characteristics of resident users' personal terminals can be used to complete the features of Area of ​​Interest (AOI). These features describe the clustering in the spatiotemporal dimension and are used to discover the regular stay of people at specific times and places.

[0068] Taking 4G / 5G resident users as the target users as an example, interest surface information is supplemented based on the spatiotemporal clustering characteristics of resident users and combined with road information from electronic maps. The location information of resident users during the day and night (time periods defined by the users themselves) is determined based on the tidal effect, including data preprocessing and user trajectory clustering analysis.

[0069] Data preprocessing mainly targets areas in electronic maps where AOIs are missing, extracting their MR location information (e.g., timestamps, latitude and longitude).

[0070] (1) Cleaning MR data: The Isolation Forest algorithm is used to detect spatiotemporal outliers, and the missing trajectories are repaired by combining the LSTM prediction model. Invalid timestamps (such as outliers) and incorrect latitude and longitude (such as geographical ranges beyond the existing communication network) are removed.

[0071] (2) Standardized time: Convert timestamps to local time and divide daytime (e.g., 09:00–17:00, adjustable) and nighttime (e.g., 22:00–06:00, adjustable).

[0072] User trajectory clustering analysis can classify location information based on time and location information. For example, the location information of users who are frequently present during the day is generally workplaces such as enterprises, institutions, office buildings, and industrial parks, while the location information of users who are frequently present at night is generally leisure and rest areas such as residential communities and shopping malls.

[0073] (1) Time segmentation: Extract user location datasets for daytime and nighttime respectively. and .

[0074] (2) Density clustering: for and Clustering using the DBSCAN algorithm, parameter settings:

[0075] in, Minimum number of samples for a radius (e.g., 500 meters, adjusted according to the actual scenario). min_samples (For example, with 50 samples, ensure that the clusters are valid regions).

[0076] (3) Output results Daytime permanent residence area Office buildings and industrial parks (high-density cluster centers, active during the day).

[0077] Nighttime permanent residence area Residential communities and shopping malls (high-density clusters, active at night).

[0078] In this embodiment, by analyzing the spatiotemporal aggregation characteristics of resident user terminals at different times, the system intelligently infers and completes the missing regional geographic entities (interest surfaces) in the electronic map, such as office buildings and residential areas. This enables automated and accurate updates of interest surface data. Utilizing the tidal effect of user groups (e.g., gathering at workplaces during the day and returning to residential areas at night), the system uses density clustering algorithms to directly deduce the precise boundaries of geographic areas from location data, overcoming the problems of slow updates and inability to identify regional functions associated with traditional satellite imagery. The generated interest surfaces not only contain geometric boundaries but also embody the functional semantics of the area (e.g., workplaces, residential areas), significantly improving the semantic richness and usability of the electronic map. Based on communication network data, it requires no manual surveying or professional mapping equipment, can operate automatically around the clock and with full coverage, and can continuously adapt to changes in urban development. It provides an efficient and economical solution for the dynamic maintenance of electronic maps under large-scale networks, offering advantages of high scalability and low cost.

[0079] In yet another exemplary embodiment, based on step 220 of the above embodiment, the method of generating the missing interest surface information of the electronic map according to the resident area and the road information of the electronic map may further include the following specific steps: The second road segment surrounding the permanent area is extracted based on the road information of the electronic map; the bounding box of the interest surface is generated based on the second road segment using the convex hull algorithm; and the interest surface information is generated based on the bounding box.

[0080] The second road segment can be obtained by constructing a road buffer and detecting spatial intersection, based on existing road network information and supplemented road information on the electronic map.

[0081] Constructing road buffer zones involves combining existing road information from electronic maps with supplemented road information to create a complete road network dataset. (See the detailed description of step 220 above), including road names, coordinates (set of line segments), and topological relationships (including intersections equipped with traffic lights), and for each road Create a buffer zone (e.g., 50 meters wide, adjustable) to represent the area affected by the road. The formula for creating a buffer zone is as follows:

[0082] in, Let be the midline point of the trajectory. Width This represents the total width of the road.

[0083] Spatial intersection detection includes each resident area or The intersection of MR location points within the clustered area and the area enclosed by the road buffer zone (including intersections equipped with traffic lights, etc.) is detected, and the bounding box is extracted. or road subset The location will be deleted at a point outside the area enclosed by the road buffer zone.

[0084] The bounding boxes of the interest surfaces are determined based on road closure, and are checked. Whether a closed loop is formed can be determined using the convex hull algorithm. Calculate the convex hull of the intersection point of the permanent area and the road (P = {p_1, p_2, ..., p_m}) to form the initial bounding box A. If the convex hull is not completely closed, supplement the intersection of adjacent roads or extend the road segment to a reasonable closure point.

[0085] In this embodiment, by integrating road buffer zones and user-resident areas, a precise bounding box for the surface of interest (AOI) is automatically generated using a convex hull algorithm, achieving highly automated and accurate generation of AOI boundaries. The road buffer zones naturally define the region's outline, and the convex hull algorithm transforms discrete road intersections into closed polygons, effectively overcoming the problem of blurred boundaries caused by relying solely on clustered point clouds. The generated geographic boundaries better reflect the physical reality of the actual road network. Using the road network dataset as a constraint for generating AOI bounding boxes ensures seamless spatial topological integration between the newly generated AOIs and existing roads, avoiding data contradictions caused by disconnections or intersections between region boundaries and roads. This helps improve the overall quality and usability of the electronic map and enhances the logical consistency between AOIs and the road network topology. By fusing road vector data with user-aggregated data and automatically fitting region boundaries using spatial algorithms, the reliance on manual interpretation and drawing is reduced. This enables rapid and batch completion of geometric information for urban functional areas, significantly reducing the cost and complexity of large-scale electronic map maintenance and improving the efficiency and reliability of map updates.

[0086] In yet another exemplary embodiment, based on step 220 of the above embodiment, the method of this embodiment may further include the following specific steps: verifying the interest surface information and outputting the verified interest surface information as target geographic feature information.

[0087] The verification process includes manual checks, which can be combined with other methods such as on-site surveys to verify the rationality of the bounding box, for example, whether it completely encloses the area of ​​interest.

[0088] The interest surface information is verified and then output in GeoJSON polygon format, containing a list of vertex latitude and longitude. GeoJSON is a format for encoding various geographic data structures. It is a geospatial information data exchange format based on JavaScript Object Notation (JSON). GeoJSON objects can represent geometry, features, or sets of features.

[0089] In this embodiment, a verification process is introduced to confirm the rationality of the automatically generated polygon of interest (POI) bounding boxes, and the results are output in a standardized format (such as GeoJSON). This improves the accuracy and reliability of the POI data. Through manual verification and other verification methods, potential boundary deviations or errors in the automatic generation algorithm (such as incompletely enclosing areas or including irrelevant empty areas) can be effectively identified and corrected, ensuring that the final output POI information highly matches the real geographic entities and guaranteeing the practical value of the data. By incorporating verification as a necessary step into the production process, a closed-loop quality control mechanism of "automatic generation + manual quality inspection" is formed, effectively reducing the risk of data errors due to algorithm limitations. This provides reliable basic data support for downstream applications (such as false alarm suppression for sensor base stations) and ensures the quality controllability of geographic data products. Using standard formats such as GeoJSON to output POI information facilitates rapid and lossless integration and visualization of the data with various geographic information systems (GIS), map platforms, and business systems (such as base station network management), significantly improving the convenience of data sharing and application, and enhancing data interoperability and integration efficiency.

[0090] Figure 3 This diagram illustrates a flowchart of another electronic map processing method provided by an embodiment of this application. This method can be executed by an electronic device, which may include a server and / or terminal devices. In other words, the method can be executed by software or hardware installed on the server and / or terminal devices, and includes the following steps: Step 310: Obtain the location data of the terminal in the communication network, the location data being used to provide the geographical location information of the terminal.

[0091] Step 320: Determine the behavioral pattern characteristics of the terminal based on the positioning data, obtain target geographic element information through the behavioral pattern characteristics, and update the electronic map according to the target geographic element information; Steps 310 and 320 can be found above. Figure 1 The detailed description of steps 110 and 120 in the illustrated embodiment, or the above Figure 2 The specific descriptions of steps 210 and 220 in the illustrated embodiment are provided, and they achieve the same technical effect. To avoid repetition, they will not be repeated here.

[0092] Step 330: Divide the updated electronic map into several spatial units; obtain simulation data of the communication network signal propagation corresponding to the updated electronic map; determine the environmental semantics of each spatial unit based on the simulation data of the communication network signal propagation and the geographic information data of the updated electronic map.

[0093] The updated electronic map is divided into several spatial units for environmental semantic reconstruction, so as to obtain an electronic map with environmental semantics.

[0094] It should be noted that environmental semantics here can refer to three-dimensional environmental semantics. When the perceived scene becomes more complex and the required accuracy of judgment is higher, three-dimensional information becomes crucial. For example, the perceived trajectory of a car driving under an overpass may overlap with the trajectory of a low-flying drone in a two-dimensional plane (2D). Only through a three-dimensional environmental semantic map can we know that an "overpass" entity exists at that location and accurately determine whether the target is under the bridge (the car, which should be filtered) or above the bridge (the drone, which should trigger an alarm). Similarly, when a drone flies between two tall buildings, its signal may be inaccurately calculated due to reflections, obstructions, etc. A three-dimensional environmental semantic map can provide the height and outline information of the buildings, allowing for a more accurate determination of whether the signal path is at line-of-sight or not, thus assisting in more precise target localization and trajectory authenticity assessment, reducing misjudgments. If the only requirement is to distinguish between ground moving targets (such as vehicles and pedestrians) and low-altitude targets (such as drones), a two-dimensional planar map with semantic attributes can meet the requirements in most cases. For example, if the location of a perceived target is projected onto the map and falls on a road, square, or other area where ground targets are allowed to exist, and its measured height is close to the ground elevation, it can be classified as a ground interference target and filtered out. The data acquisition, processing, and calculation costs of two-dimensional maps are much lower than those of three-dimensional maps, making them easier to implement and deploy on a large scale.

[0095] Based on the latest electronic map (5m satellite map) with complete road information and interest surface bounding box location information, a 3D environmental semantic map is reconstructed. Each spatial unit can be regarded as a 3D environmental semantic grid, with a size of, for example, 5m*5m*5m (length*width*height, the values ​​can be adjusted according to requirements), including lane elevation information, lane border position, lane and AOI border semantic attribution, etc.

[0096] The simulation data for communication network signal propagation can be obtained through diffraction calculations of the reflection paths of the inductive base station signal at line-of-sight (LOS), near-LOS, and non-line-of-sight (NLOS). The attribute information of LOS, Near-LOS, and NLOS can be used to characterize signal propagation characteristics. Line-of-sight (LOS) propagation is a communication method where radio waves transmit directly between the transmitting and receiving points, typically over distances of 20-50 kilometers, primarily used for VHF and microwave communications. Line-of-sight (LOS) refers to direct signal transmission without obstruction. Near-LOS refers to the presence of some obstacles (such as trees or buildings) but not complete obstruction of signal transmission. Non-line-of-sight (NLOS) refers to complete signal obstruction, requiring transmission through reflection or diffraction.

[0097] The geographic information data in electronic maps can include vector layers and cluster layers. Vector layers are the core component of electronic maps used to represent geometric objects such as points, lines, and polygons. Their data is stored in coordinate chains (x, y, z), supporting precise descriptions of the spatial location and topological relationships of geographic entities. For example, road and building outlines can be presented as polygons or lines using vector data. Cluster layers use aggregation algorithms to visualize dense point features (e.g., POIs, buildings) in a clustered or dispersed manner, reducing map load and improving rendering efficiency. For example, urban heatmaps or massive point data can dynamically adjust display density through hierarchical clustering styles.

[0098] Based on the diffraction data obtained from the LOS, Near-LOS, and NLOS reflection paths of the inductive base station signals, and combined with information such as vector layers and feature layers in the electronic map, a stereo semantic raster is constructed based on several spatial units divided by the electronic map. This raster generates information including urban roads, buildings, bridges, open areas, water bodies, vegetation, playgrounds, etc., as shown in the table below:

[0099] In this embodiment, by discretizing the updated electronic map into spatial units and fusing communication signal propagation simulation data and geographic information data, each unit is given precise environmental semantics, enabling refined and three-dimensional reconstruction of environmental semantic information. By introducing signal propagation simulation (e.g., LOS / NLOS judgment) and three-dimensional grid division, the map can not only express the planar outline of ground features but also depict their spatial three-dimensional structure and electromagnetic propagation characteristics. This allows for accurate determination of target height in complex urban scenarios (e.g., overpasses, between buildings), effectively distinguishing between ground vehicles and low-altitude drones. The multi-dimensional semantic grid obtained by fusing geographic elements and signal propagation characteristics provides rich prior knowledge for the sensing base station, enabling it to directly filter false alarms based on spatial unit semantics (e.g., "NLOS area," "building interior") or analyze fuzzy trajectories by combining three-dimensional information, significantly reducing the probability of misjudgment and helping to improve the perception system's ability to recognize and resist interference in complex environments.

[0100] Furthermore, the method in this embodiment can also support the generation of semantic maps of different granularities (e.g., low-cost two-dimensional semantic layers or high-precision three-dimensional semantic grids). The implementation method can be flexibly selected according to actual needs (such as regional importance and computing resources). While ensuring the perception accuracy of key areas, it controls the overall deployment and computing overhead, has good scalability and practicality, and can balance accuracy and cost in different application scenarios.

[0101] In yet another exemplary embodiment, based on step 220 of the above embodiment, this embodiment further includes the following specific steps: The system acquires sensing data obtained from sampling the monitoring space over a continuous time period. The sensing data includes trajectory sampling points generated by moving targets. For each trajectory sampling point, a judgment is performed. If the kinematic features of the trajectory sampling point match a preset motion model and the altitude of the trajectory sampling point is within a preset range, the trajectory sampling point is marked as an interference sampling point. For each spatial unit, statistics are performed. If the proportion of interference sampling points exceeds a specific threshold, the spatial unit is marked as an interference region. Trajectory sampling points located within the interference region are filtered.

[0102] Based on the above embodiments, electronic maps can be updated and environmental semantics reconstructed. The resulting three-dimensional environmental semantic map is loaded onto a sensing base station and can be used to achieve ground lattice lobe identification and false alarm suppression. The specific implementation method mainly includes the following steps: (1) Data input includes geographic information data from electronic maps (e.g., 5m satellite maps) and high-precision engineering parameter information from the sensing base station. The geographic information data includes vector layers of various land features such as roads, buildings, bridges, water bodies, and vegetation. The high-precision engineering parameter information includes parameters such as its precise geographical location, antenna azimuth angle, and downtilt angle.

[0103] (2) Semantic grid division: The size of each three-dimensional environment semantic grid is set according to the requirements. The semantic grid includes lane elevation information, lane border position, lane and AOI border semantic attribution, etc.

[0104] (3) Wireless signal propagation simulation: Based on the engineering parameters of the base station and three-dimensional geographic environment data, the propagation process of the sensing signal in space is simulated. By calculating the direct, reflection and diffraction effects of the signal, the base station coverage area is divided into regions with different signal propagation characteristics, mainly including line-of-sight region, near-line-of-sight region and non-line-of-sight region, which adds a layer of "signal propagation semantics" to the physical environment.

[0105] (4) Multi-source layer fusion and 3D semantic grid division: The signal propagation characteristic layer obtained from the above simulation is superimposed and fused with the geographic feature layers in the original electronic map to generate and output a 3D environmental semantic map. Each 3D grid is assigned a set of fused semantic attribute labels, such as the type of land cover corresponding to its geographical location (e.g., road, building, etc.) and the signal propagation status (e.g., LOS, NLOS, etc.).

[0106] (5) Once the three-dimensional environmental semantic map is loaded into the sensing base station, it can provide crucial environmental context knowledge for accurate target recognition and false alarm suppression.

[0107] The sensing data can include the sensing trajectories of drones, vehicles, pedestrians, and birds identified by the sensing base station from the ground grating lobe. It mainly suppresses false alarms of moving targets such as vehicles / pedestrians on the road and pedestrians within the AOI frame (vehicles / pedestrians account for the largest proportion). Generally, it can be divided into the following situations: false alarm trajectories appearing inside buildings and within NLOS areas are directly removed; for 4G / 5G vehicle users or pedestrians on the road, based on kinematic features (vehicle speed, acceleration, rate of change of direction, trajectory curvature) and altitude (Height), a deep neural network is used to learn false alarms of moving targets such as vehicles and pedestrians. If the proportion of clutter points in the three-dimensional semantic grid exceeds the classification threshold for false alarm learning, the grid can be regarded as a clutter grid, and its corresponding trajectory data can be regarded as a false alarm trajectory, which will not be reported in the monitoring platform.

[0108] This study uses a deep neural network to learn false alarms from moving clutter such as vehicles and pedestrians. Inputs include: the sensing trajectories (timestamp, latitude and longitude) detected by the integrated sensing base station, with one sampling point generated every 640ms, including trajectories generated by moving targets such as drones, vehicles, and pedestrians; and a 3D environmental semantic map, including the environmental semantic attribution, latitude and longitude, and altitude of each grid cell. Outputs include: if a continuous sampling point matches the four-dimensional features of the kinematic trajectory of a vehicle or pedestrian, and its altitude is above the road altitude but below the sensing base station altitude, it is identified as a clutter sampling point; if the proportion of clutter sampling points in the grid (clutter sampling points / total sampling points) is greater than the clutter false alarm learning classification threshold (75%, adjustable as required), the grid can be considered a clutter grid where vehicles and pedestrians gather, and the generated sensing trajectories are considered false alarm trajectories and directly deleted from the base station monitoring platform.

[0109] In this embodiment, a three-dimensional environmental semantic map is constructed, endowing the base station with a deep understanding of physical space. This map not only includes static geographic information such as roads and buildings, but also integrates the key dimension of signal propagation characteristics (LOS / NLOS), enabling the base station to distinguish seemingly similar but actually different scenarios. For example, by recognizing the three-dimensional structure, it can accurately determine whether the target is a vehicle located under an overpass or a drone above the bridge. False trajectories inside buildings can be directly eliminated using NLOS tags, transforming environmental perception from passive perception to active understanding. A dual intelligent suppression mechanism is introduced: rapid filtering based on the semantic map directly eliminates trajectories falling into clearly invalid areas such as inside buildings and NLOS regions, achieving efficient initial screening; and deep learning models analyze the kinematic characteristics (velocity, acceleration, curvature, etc.) and altitude information of the trajectories to distinguish drones from ground interference in real time. By statistically analyzing the clutter ratio within the grid and setting an adaptive threshold, high-frequency areas of vehicles and pedestrians can be dynamically identified, and similar trajectories within these areas can be batch-suppressed, significantly improving the accuracy and robustness of false alarm identification. By endowing base stations with environmental context understanding capabilities, false alarms caused by environmental misunderstandings are fundamentally reduced, making it particularly suitable for complex urban scenarios. The automated learning mechanism reduces reliance on manual rule configuration and frequent optimization, enabling autonomous operation and maintenance of the system. It supports flexible selection of 2D / 3D map accuracy according to scenario requirements, balancing computational costs and perception needs, and has good scalability.

[0110] In yet another exemplary embodiment, the method of this embodiment may further include the following specific steps: regularly updating and iterating the electronic map, including: periodically completing the road information and interest surface bounding boxes in the electronic map based on 4G / 5G vehicle users and all users, and updating the three-dimensional environmental semantic map in a timely manner.

[0111] In this embodiment, by introducing a normalized update and iteration mechanism, road and interest surface information is automatically supplemented using continuously acquired communication network user data. This enables the three-dimensional environmental semantic map to dynamically adapt to urban development and changes, effectively overcoming the accuracy decay problem caused by the lag in updates of traditional static maps. This provides a continuous and reliable environmental cognition foundation for the integrated sensing base station, ensuring the long-term effectiveness and stability of the false alarm suppression effect.

[0112] Corresponding to the electronic map processing method provided in the above embodiments, based on the same technical concept, this application also provides an electronic map processing system. The system includes a data module and an update module.

[0113] The data module is used to acquire the location data of the terminal in the communication network; the update module is used to determine the behavioral pattern characteristics of the terminal based on the location data, obtain target geographic element information through the behavioral pattern characteristics, and update the electronic map according to the target geographic element information.

[0114] It should be noted that the electronic map processing system and the electronic map processing method provided in this application embodiment are based on the same application concept. Therefore, the specific implementation of this embodiment can refer to the implementation of the aforementioned electronic map processing method, and the repeated parts will not be described again.

[0115] Corresponding to the electronic map processing method provided in the above embodiments, based on the same technical concept, this application also provides a communication device. See Figure 4 The communication device 400 includes a communication unit 410, a sensing unit 420, and a processing unit 430.

[0116] The communication unit 410 is configured to wirelessly communicate with one or more terminals in a communication network and acquire the positioning data of the terminals; the sensing unit 420 is configured to transmit sensing signals and receive their echo signals to acquire target trajectory data within the monitoring space; and the processing unit 430 is connected to the communication unit and the sensing unit and is configured to update the electronic map according to the electronic map processing method described in the above embodiments, and use the updated electronic map to process the target trajectory data.

[0117] It should be noted that the processing unit provided in this application embodiment and the electronic map processing method provided in this application embodiment are based on the same application concept. Therefore, the specific implementation of this embodiment can refer to the implementation of the aforementioned electronic map processing method, and the repeated parts will not be described again.

[0118] The communication device provided in this embodiment, applied to an integrated sensing base station, can reconstruct missing roads, residential areas, office buildings, industrial parks, and other AOIs in the 5m satellite map based on the MR location information of 4G / 5G new energy vehicle users and all permanent users of the communication network, using AI algorithms and related model tools. This enables efficient, accurate, and real-time reconstruction of a 3D environmental semantic map, ultimately reducing false alarms. With the increasing penetration rate of new energy vehicles and the widespread adoption of assisted driving, these vehicles can operate in many map-free scenarios, typically in suburban and rural areas not currently mapped by the 5m satellite imagery. This method further supplements the 5m electronic map with road information, continuously building a 3D semantic map across the entire network. By supplementing missing roads and properties in the satellite map based on the MR location information of new energy vehicle users and permanent users, and then optimizing the sensing 3D semantic map, this method can complete the electronic map using anonymized user data from the communication network at a lower cost, significantly reducing the false alarm probability of the sensing base station. In terms of communication equipment capabilities, it reduces the system's false alarm probability and saves costs.

[0119] Corresponding to the electronic map processing method provided in the above embodiments, based on the same technical concept, this application also provides an electronic device for executing the above-described electronic map processing method. Figure 5 To illustrate the structure of an electronic device according to various embodiments of this application, as shown in the following diagrams... Figure 5 As shown. Electronic device 500 can vary considerably due to differences in configuration or performance, and may include one or more processors 510 and memory 520. Memory 520 may store one or more application programs or data. Memory 520 may be temporary or persistent storage. The application programs stored in memory 520 may include one or more modules (not shown), each module may include a series of computer-executable instructions for the electronic device. Furthermore, processor 510 may be configured to communicate with memory 520 and execute the series of computer-executable instructions stored in memory 520 on the electronic device.

[0120] This application also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions, which, when executed by a computer, implement the steps of the electronic map processing method described above.

[0121] Those skilled in the art will understand that embodiments of this application can be provided as methods, apparatus, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0122] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0123] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0124] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0125] In a typical configuration, an electronic device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0126] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0127] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0128] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0129] Those skilled in the art will understand that embodiments of this application can be provided as methods, apparatus, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0130] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for processing electronic maps, characterized in that, The method includes the following steps: Obtain the location data of the terminal in the communication network; Based on the location data, the behavioral pattern characteristics of the terminal are determined, target geographic element information is obtained through the behavioral pattern characteristics, and the electronic map is updated according to the target geographic element information.

2. The method according to claim 1, characterized in that, The process of determining the terminal's behavioral pattern characteristics based on the location data, obtaining target geographic element information through the behavioral pattern characteristics, and updating the electronic map based on the target geographic element information includes the following steps: Trajectory data from the vehicle terminal is selected from the positioning data. Based on the kinematic features of the trajectory data, road boundary information missing from the electronic map is generated. The road boundary information is then used as the target geographic feature information and topologically connected to the existing road network in the electronic map.

3. The method according to claim 2, characterized in that, The step of selecting trajectory data from the vehicle terminal from the positioning data and generating the missing road boundary information of the electronic map based on the kinematic features of the trajectory data includes the following steps: Calculate the kinematic characteristics of the trajectory data; Based on the density clustering algorithm and the calculated kinematic features, the trajectory points in the trajectory data are clustered, and the missing candidate road areas in the electronic map are identified based on the clustering results. The identified candidate road areas are filtered for directional consistency to determine continuous first road segments; The road boundary information is generated based on the trajectory point set of the first road segment using the convex hull algorithm.

4. The method according to claim 1, characterized in that, The step of determining the terminal's behavioral pattern characteristics based on the positioning data, obtaining target geographic element information through the behavioral pattern characteristics, and updating the electronic map based on the target geographic element information further includes the following steps: Based on the time information of the location data, the location data of the target user in the communication network is divided into datasets for different time periods, and the datasets for each time period are clustered to determine the target user's permanent residence area in different time periods. Based on the permanent residence area and the road information of the electronic map, interest surface information missing from the electronic map is generated, and the interest surface information is used as target geographic feature information.

5. The method according to claim 4, characterized in that, The step of generating the missing interest surface information of the electronic map based on the resident area and the road information of the electronic map includes the following steps: Extract the second road segment surrounding the permanent area based on the road information of the electronic map; The bounding box of the interest surface is generated based on the second road segment using the convex hull algorithm, and the interest surface information is generated based on the bounding box.

6. The method according to claim 1, characterized in that, The method further includes the following steps: The updated electronic map is divided into several spatial units; Obtain simulation data of the communication network signal propagation corresponding to the updated electronic map; The environmental semantics of each spatial unit are determined based on the simulation data of the communication network signal propagation and the geographic information data of the updated electronic map.

7. The method according to claim 6, characterized in that, The method further includes the following steps: Acquire sensing data obtained by sampling the monitored space over a continuous time period, wherein the sensing data includes trajectory sampling points generated by a moving target; For each trajectory sampling point, a judgment is performed. If the kinematic features of the trajectory sampling point match the preset motion model and the altitude of the trajectory sampling point is within the preset range, the trajectory sampling point is marked as an interference sampling point. Statistical analysis is performed on each of the spatial units, and if the proportion of interfering sampling points exceeds a specific threshold, the spatial unit is marked as an interfering region. The trajectory sampling points located within the interference area are filtered.

8. A communication device, characterized in that, Including the following: The communication unit is configured to wirelessly communicate with one or more terminals in a communication network and acquire the location data of the terminals; The sensing unit is configured to transmit sensing signals and receive their echo signals in order to acquire target trajectory data within the monitoring space. as well as The processing unit is connected to the communication unit and the sensing unit and is configured to update the electronic map according to the processing method of any one of claims 1 to 7, and to process the target trajectory data using the updated electronic map.

9. An electronic device, characterized in that, It includes a processor, a memory, a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the method as described in any one of claims 1 to 7.

10. A computer program product, characterized in that, The computer program product includes a computer program stored on a non-transitory computer-readable storage medium, the computer program including program instructions that, when executed by a computer, implement the steps of the method as described in any one of claims 1 to 7.

Images