Communication network optimization system and method based on spatiotemporal data fusion

Abstract

The present invention provides a communication network optimization system and method based on spatiotemporal data fusion, which relates to the technical field of real-time data stream processing. K potential blind spot sequences are output by performing base station coverage analysis on the predicted driving trajectory; K potential blind spot sequences are analyzed for driving characteristics, and K channel switching request sequences are output to construct a communication network optimization library; according to the real-time spatiotemporal data of the target vehicle, the switching instructions are dispatched from the communication network optimization library to dynamically update the communication link of the target vehicle. The technical problem that the data processing response delay of the communication link switching analysis in the high-speed mobile scenario in the prior art leads to weak stability and reliability of the communication network coverage in the high-speed mobile scenario, and a high risk of communication interruption is solved. The technical effect of improving the data analysis efficiency of dynamic switching of communication links in high-speed mobile communication scenarios, enhancing the reliability and coverage performance of the communication network, and effectively reducing the risk of communication interruption is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of real-time data stream processing, and in particular to a communication network optimization system and method based on spatiotemporal data fusion. Background Art

[0002] In high-speed mobile scenarios, existing communication technologies face many challenges, especially in data processing and storage optimization. First, high-speed movement leads to a significant increase in Doppler shift and serious signal frequency deviation, which not only affects signal synchronization and reliability, but also increases the complexity of data processing. Existing communication systems often lack efficient computing models and optimization algorithms when dealing with these dynamic changes, resulting in low data processing efficiency.

[0003] Secondly, when vehicles quickly cross different base station coverage areas, base station switching is frequent, which can easily lead to signal interruption or jamming. This process involves a large amount of data transmission and storage operations. The existing system has deficiencies in data storage and management and cannot efficiently handle frequent switching requests, further exacerbating the risk of communication interruption.

[0004] In addition, complex terrain and buildings will further aggravate signal attenuation and reduce communication stability. This not only affects signal quality, but also increases the difficulty of data processing. These factors together lead to insufficient coverage stability of communication networks in high-speed mobile scenarios, and there is a high risk of communication interruption.

[0005] In summary, the prior art has technical problems such as data processing response delay in communication link switching analysis under high-speed mobile scenarios, which leads to weak communication network coverage stability and reliability and high risk of communication interruption under high-speed mobile scenarios. Summary of the invention

[0006] The present invention provides a communication network optimization system and method based on spatiotemporal data fusion, which is used to solve the technical problems in the prior art of data processing response delay in communication link switching analysis under high-speed mobile scenarios, resulting in weak communication network coverage stability and reliability and high risk of communication interruption under high-speed mobile scenarios.

[0007] In view of the above problems, the present invention provides a communication network optimization system and method based on spatiotemporal data fusion.

[0008] According to a first aspect of the present invention, a communication network optimization system based on spatiotemporal data fusion is provided, the system comprising: a state correction execution module, which is used to output a real-time corrected state vector by fusing and processing the real-time three-dimensional position and real-time inertial data of a target vehicle; a driving trajectory prediction module, which is used to perform driving trajectory fitting in a road network topology according to the real-time three-dimensional position, and output a predicted driving trajectory; a signal coverage analysis module, which is used to pre-calculate the base station coverage strength of the predicted driving trajectory and output signal coverage distribution characteristics; a signal coverage correction module, which is used to correct the signal coverage distribution characteristics based on scene elements and obtain K potential blind spot sequences; a channel switching analysis module, which is used to perform driving characteristic analysis on the K potential blind spot sequences and output K channel switching request sequences; a communication network optimization module, which is used to associate and store the K potential blind spot sequences and the K channel switching request sequences to obtain a communication network optimization library; a communication optimization execution module, which is used to dynamically update the communication link of the target vehicle by scheduling switching instructions from the communication network optimization library according to the real-time spatiotemporal data of the target vehicle.

[0009] The second aspect of the present invention provides a communication network optimization method based on spatiotemporal data fusion, the method comprising: outputting a real-time corrected state vector by fusing and processing the real-time three-dimensional position and real-time inertial data of the target vehicle; performing driving trajectory fitting in the road network topology according to the real-time three-dimensional position, and outputting a predicted driving trajectory; pre-calculating the base station coverage intensity of the predicted driving trajectory, and outputting signal coverage distribution characteristics; correcting the signal coverage distribution characteristics based on scene elements to obtain K potential blind spot sequences; performing driving characteristic analysis on the K potential blind spot sequences, and outputting K channel switching request sequences; associatively storing the K potential blind spot sequences and the K channel switching request sequences to obtain a communication network optimization library; and scheduling switching instructions from the communication network optimization library according to the real-time spatiotemporal data of the target vehicle to dynamically update the communication link of the target vehicle.

[0010] One or more technical solutions provided in the present invention have at least the following technical effects or advantages:

[0011] The system provided by the embodiment of the present invention is used to output a real-time corrected state vector by fusing the real-time three-dimensional position and real-time inertial data of the target vehicle through a state correction execution module; a driving trajectory prediction module is used to fit the driving trajectory in the road network topology according to the real-time three-dimensional position and output the predicted driving trajectory; a signal coverage analysis module is used to pre-calculate the base station coverage intensity of the predicted driving trajectory and output the signal coverage distribution characteristics; a signal coverage correction module is used to correct the signal coverage distribution characteristics based on scene elements to obtain K potential blind area sequences; a channel switching analysis module is used to perform driving characteristic analysis on the K potential blind area sequences and output K channel switching request sequences; a communication network optimization module is used to associate and store the K potential blind area sequences and K channel switching request sequences to obtain a communication network optimization library; a communication optimization execution module is used to dispatch the switching instructions from the communication network optimization library according to the real-time spatiotemporal data of the target vehicle to dynamically update the communication link of the target vehicle. The technical effect of improving the data analysis efficiency of dynamic switching of communication links in high-speed mobile communication scenarios, enhancing the reliability and coverage performance of the communication network, and effectively reducing the risk of communication interruption is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 A schematic diagram of the structure of a communication network optimization system based on spatiotemporal data fusion provided by the present invention.

[0013] Figure 2 A schematic flow chart of the communication network optimization method based on spatiotemporal data fusion provided by the present invention.

[0014] Explanation of the accompanying drawings: state correction execution module 11, driving trajectory prediction module 12, signal coverage analysis module 13, signal coverage correction module 14, channel switching analysis module 15, communication network optimization module 16, communication optimization execution module 17. DETAILED DESCRIPTION

[0015] The present invention provides a communication network optimization system and method based on spatiotemporal data fusion, which is used to solve the technical problem that the data processing response delay of the communication link switching analysis in the high-speed mobile scene in the prior art leads to the weak stability and reliability of the communication network coverage in the high-speed mobile scene, and the high risk of communication interruption. The technical effect of improving the data analysis efficiency of dynamic switching of communication links in the high-speed mobile communication scene, enhancing the reliability and coverage performance of the communication network, and effectively reducing the risk of communication interruption is achieved.

[0016] Below, the technical solutions in the present invention will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments of the present invention. It should be understood that the present invention is not limited to the example embodiments described herein. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. It should also be noted that, for the convenience of description, only the parts related to the present invention are shown in the accompanying drawings, rather than all of them.

[0017] Embodiment 1, as Figure 1 As shown, the present invention provides a communication network optimization system based on spatiotemporal data fusion, wherein the system comprises:

[0018] The state correction execution module 11 is used to output a real-time correction state vector by fusing and processing the real-time three-dimensional position and real-time inertial data of the target vehicle.

[0019] The driving trajectory prediction module 12 is used to perform driving trajectory fitting in the road network topology according to the real-time three-dimensional position and output a predicted driving trajectory.

[0020] The signal coverage analysis module 13 is used to pre-calculate the base station coverage strength for the predicted driving trajectory and output signal coverage distribution characteristics.

[0021] The signal coverage correction module 14 is used to correct the signal coverage distribution characteristics based on scene elements to obtain K potential blind area sequences.

[0022] The channel switching analysis module 15 is used to perform driving characteristic analysis on the K potential blind spot sequences and output K channel switching request sequences.

[0023] The communication network optimization module 16 is used to associate and store the K potential blind area sequences and the K channel switching request sequences to obtain a communication network optimization library.

[0024] The communication optimization execution module 17 is used to dynamically update the communication link of the target vehicle by scheduling switching instructions from the communication network optimization library according to the real-time spatiotemporal data of the target vehicle.

[0025] In one embodiment, the state correction execution module 11 is used to:

[0026] The real-time three-dimensional position of the target vehicle is obtained by interacting with the on-board GPS; after calibrating the IMU sensor, the IMU sensor is operated to collect the real-time inertial data, wherein the real-time inertial data includes real-time acceleration and real-time angular velocity; the real-time driving state parameters are read through the OBD interface, wherein the real-time driving state parameters include real-time speed and real-time driving direction; the real-time three-dimensional position, real-time inertial data and real-time driving state parameters are fused and processed by applying the Kalman filter algorithm, and the real-time corrected state vector is output.

[0027] In one embodiment, the driving trajectory prediction module 12 is used to:

[0028] The spatiotemporal optimization range is matched according to the real-time driving state parameters; the real-time three-dimensional position is used as the starting point, and the associated road network is called according to the spatiotemporal optimization range, and the real-time road network topology is output; the real-time corrected state vector is used as the direction constraint, and the driving trajectory is fitted in the real-time road network topology, and the predicted driving trajectory is output, wherein the predicted driving trajectory includes K alternative driving trajectories.

[0029] In one embodiment, the driving trajectory prediction module 12 is used to:

[0030] The control point demand is matched according to the real-time driving state parameters to obtain a control point quantity constraint; the real-time corrected state vector is used as a direction constraint to locate a road network control point array that satisfies the control point quantity constraint in the real-time road network topology; the real-time driving state parameters are used to predict future trajectories in the road network control point array, and a control point trajectory array is output; the real-time road network topology is used as a boundary constraint to smooth the driving trajectory of the control point trajectory array to obtain the predicted driving trajectory.

[0031] In one embodiment, the signal coverage analysis module 13 is used to:

[0032] The first candidate driving trajectory is decomposed to obtain a first trajectory point sequence; with the base station coverage radius as a constraint, the first trajectory point sequence is searched for covering base stations to locate the first covering base station sequence; signal propagation prediction is performed on the first covering base station sequence, and the base station signal stability is judged according to the prediction result to obtain a first signal coverage strength sequence; by analogy, the base station coverage strength is pre-calculated for the K candidate driving trajectories, and K signal coverage strength sequences are output; the K signal coverage strength sequences constitute the signal coverage distribution characteristics.

[0033] In one embodiment, the signal coverage analysis module 13 is used to:

[0034] Perform signal propagation prediction on the first coverage base station sequence to obtain W groups of signal strength characteristics of the W first trajectory points in the first trajectory point sequence; perform base station aggregation on the W groups of signal strength characteristics to obtain H initial service base stations; perform base station signal stability evaluation based on the W groups of signal strength characteristics, and screen the H initial service base stations to obtain a first preferred service base station and H-1 first alternative service base stations based on the evaluation results; use the first preferred service base station as a screening condition, and screen the W groups of signal strength characteristics to obtain W trajectory point signal coverage strengths, and the W trajectory point signal coverage strengths constitute the first signal coverage strength sequence.

[0035] In one embodiment, the signal coverage correction module 14 is used to:

[0036] The K scene element sequences of the K alternative driving trajectories are called online; the K scene element sequences are used to correct the K signal coverage strength sequences, and K signal coverage correction sequences are output; and the K potential blind spot sequences are located in the K alternative driving trajectories according to the K signal coverage correction sequences.

[0037] In one embodiment, the channel switching analysis module 15 is used to:

[0038] A first potential blind spot sequence is used to traverse the W first trajectory points in the first trajectory point sequence to locate L first blind spot points; the L first blind spot points are used as coverage position screening conditions, the H-1 first candidate service base stations are used as base station screening conditions, and L groups of blind spot signal coverage strengths are obtained from the W groups of signal strength characteristics; the L groups of blind spot signal coverage strengths are serialized within the group, and switching base stations are selected according to the sorting results to obtain L base station switching requests; the L base station switching requests and the L first blind spot points are associated and stored to obtain a first channel switching request sequence; and by analogy, the K potential blind spot sequences are subjected to driving characteristic analysis to output the K channel switching request sequences.

[0039] In one embodiment, the driving trajectory prediction module 12 is used to:

[0040] The real-time driving state parameter is used to traverse the driving trajectory change table to obtain a trajectory change response scale; the road network control point array is used as a traffic position retrieval condition, and the trajectory change response scale is used as a traffic time retrieval condition to perform a traffic data networking call to obtain M groups of historical driving trajectories of M road network control points in the road network control point array; M groups of updated driving trajectories are obtained by performing trajectory overlap frequency analysis on the M groups of historical driving trajectories; and the M groups of updated driving trajectories are mapped and synchronized to the real-time road network topology according to the road network control point array to obtain the control point trajectory array.

[0041] Embodiment 2, as Figure 2 As shown, the present invention provides a communication network optimization method based on spatiotemporal data fusion, the method comprising:

[0042] A100: Outputs a real-time correction state vector by fusing and processing the real-time 3D position and real-time inertial data of the target vehicle.

[0043] In one embodiment, by fusing and processing the real-time three-dimensional position and real-time inertial data of the target vehicle, a real-time correction state vector is outputted. Step A100 of the method provided by the present invention includes:

[0044] A110: Interact with the vehicle-mounted GPS to obtain the real-time three-dimensional position of the target vehicle.

[0045] A120: After calibrating the IMU sensor, run the IMU sensor to collect the real-time inertial data, wherein the real-time inertial data includes real-time acceleration and real-time angular velocity.

[0046] A130: Read real-time driving status parameters through the OBD interface, wherein the real-time driving status parameters include real-time speed and real-time driving direction.

[0047] A140: Apply the Kalman filter algorithm to fuse and process the real-time three-dimensional position, real-time inertial data and real-time driving state parameters, and output the real-time corrected state vector.

[0048] Specifically, in this embodiment, by interacting with the on-board GPS module of the target vehicle, the real-time three-dimensional position information of the target vehicle is obtained, and the real-time three-dimensional position information includes longitude, latitude and elevation, which can accurately reflect the position of the vehicle in space.

[0049] Before running the IMU sensor, it is first calibrated to ensure the measurement accuracy of the sensor. The IMU sensor specifically includes an accelerometer and a gyroscope, which can measure the real-time acceleration and angular velocity of the vehicle. The accelerometer is used to measure the acceleration of the vehicle in three axes, while the gyroscope is used to measure the angular velocity of the vehicle. The real-time inertial data reflects the motion state of the vehicle, including speed changes and direction changes.

[0050] Based on the characteristic that the OBD interface can provide detailed information reflecting the movement state of the vehicle, this embodiment reads the real-time driving state parameters including the real-time speed and the real-time driving direction through the OBD interface of the vehicle. Among them, the real-time speed reflects the current driving speed of the vehicle, and the real-time driving direction indicates the driving direction of the vehicle.

[0051] The real-time three-dimensional position, real-time inertial data and real-time driving state parameters are input into the Kalman filter algorithm for fusion processing. The specific fusion processing process is as follows:

[0052] First, the state vector, covariance matrix, and covariance matrix of process noise and observation noise of the Kalman filter are initialized. At each time step, the prediction step is performed to predict the state vector at the current moment using the state transition model and update the covariance matrix of the predicted state.

[0053] Subsequently, an update step is performed to calculate the Kalman gain, calculate the observation residual according to the observation model, correct the state vector using the Kalman gain and the observation residual, and update the covariance matrix.

[0054] Through this process, the Kalman filter algorithm comprehensively processes the three-dimensional position information provided by the GPS, the acceleration and angular velocity information provided by the IMU sensor, and the driving status parameters provided by the OBD interface, and finally outputs the real-time corrected state vector containing the vehicle's high-precision position, speed, and direction information, providing accurate input data for subsequent communication network optimization.

[0055] This embodiment achieves the technical effect of effectively improving the vehicle positioning accuracy and motion state estimation accuracy by fusing and processing the real-time three-dimensional position and real-time inertial data of the target vehicle and outputting a real-time correction state vector.

[0056] A200: Perform driving trajectory fitting in the road network topology according to the real-time three-dimensional position, and output a predicted driving trajectory.

[0057] In one embodiment, a driving trajectory is fitted in a road network topology according to the real-time three-dimensional position, and a predicted driving trajectory is output. Step A200 of the method provided by the present invention includes:

[0058] A210: Match the spatiotemporal optimization range according to the real-time driving status parameters.

[0059] A220: Taking the real-time three-dimensional position as the starting point, performing associated road network calls according to the spatiotemporal optimization range, and outputting real-time road network topology.

[0060] A230: Using the real-time corrected state vector as a direction constraint, performing driving trajectory fitting in the real-time road network topology, and outputting the predicted driving trajectory, wherein the predicted driving trajectory includes K alternative driving trajectories.

[0061] In one embodiment, the real-time corrected state vector is used as a direction constraint, a driving trajectory is fitted in the real-time road network topology, and the predicted driving trajectory is output. Step A230 of the method provided by the present invention includes:

[0062] A231: Match the control point demand according to the real-time driving status parameters to obtain the control point quantity constraint.

[0063] A232: Using the real-time corrected state vector as a direction constraint, locate a road network control point array that satisfies the control point quantity constraint in the real-time road network topology.

[0064] A233: Using the real-time driving state parameters, perform future trajectory prediction on the road network control point array and output a control point trajectory array.

[0065] A234: Using the real-time road network topology as a boundary constraint, smoothing the driving trajectory of the control point trajectory array to obtain the predicted driving trajectory.

[0066] In one embodiment, the real-time driving state parameters are used to perform future trajectory prediction on the road network control point array, and the control point trajectory array is output. Step A233 of the method provided by the present invention includes:

[0067] A2331: Use the real-time driving state parameters to traverse the driving trajectory change table to obtain a trajectory change response scale.

[0068] A2332: Using the road network control point array as a traffic position retrieval condition, using the trajectory change response scale as a traffic time retrieval condition, and performing a traffic data network call to obtain M groups of historical driving trajectories of the M road network control points in the road network control point array.

[0069] A2333: Analyze the overlap frequency of the M groups of historical driving trajectories to obtain M groups of updated driving trajectories.

[0070] A2334: Synchronize the M groups of updated driving trajectory mappings to the real-time road network topology according to the road network control point array to obtain the control point trajectory array.

[0071] Specifically, in this embodiment, according to the real-time driving state parameters (such as speed and driving direction) of the target vehicle, a matching spatiotemporal optimization range is determined. The spatiotemporal optimization range refers to the spatial range in which the vehicle may travel in the future, which is used to limit the calculation range of trajectory prediction and improve calculation efficiency and accuracy.

[0072] Taking the real-time 3D position of the target vehicle as the starting point, combined with the spatiotemporal optimization range, the road network topology information in the high-precision map is called. Through interaction with the map service, the real-time road network data within the range is obtained, including road type, number of lanes, traffic signs, signal light locations, and road connection relationships, etc. Then, the real-time road network topology is constructed based on the real-time road network data to provide basic data for trajectory fitting.

[0073] It should be understood that in this embodiment, the control points are key points for trajectory fitting, and their number and position directly affect the smoothness and accuracy of the trajectory. Based on this, this embodiment uses the real-time driving state parameter to traverse the control point demand quantity table, matching the control point quantity constraint whose speed and acceleration in the real-time driving state parameter fall into the sample speed interval and sample acceleration interval, and the control point quantity constraint meets the trajectory fitting accuracy requirement.

[0074] In the real-time road network topology, according to the direction constraint of the real-time corrected state vector, a road network control point array that satisfies the control point quantity constraint is located. The array includes a total of M road network control points. These control points are located in the expected driving direction of the vehicle and conform to the geometric constraints of the road network. Adjacent control points on the same road have the same spacing distance. The road network control point array provides a key reference point for trajectory fitting.

[0075] The real-time driving state parameters (such as current speed and acceleration) of the target vehicle are used to query the driving trajectory change table, which records the typical response time of the driver to make driving control adjustments under different driving states. By matching the current driving state, the corresponding trajectory change response scale is determined. The trajectory change response scale reflects the driver's reaction time to driving operations (such as acceleration, deceleration, and steering) under the current speed and acceleration, providing a time benchmark for subsequent trajectory prediction.

[0076] The road network control point array is used as the geospatial retrieval condition, and the trajectory change response scale is combined as the time retrieval condition. The traffic data service is called through the network to obtain the historical driving trajectory data matching these conditions. Specifically, the historical driving trajectories of other vehicles passing through the M road network control points within a similar location and time range are retrieved from the traffic database, totaling M groups of historical driving trajectories.

[0077] It should be understood that the M groups of historical driving trajectories reflect the actual driving paths of the vehicles at the M road network control points under similar driving conditions and response times, providing a reference for predicting the future trajectory of the target vehicle.

[0078] The M sets of historical driving trajectories obtained are analyzed, and the overlap frequencies between the trajectories are counted. Through this analysis, trajectory patterns that frequently appear under similar driving conditions are screened out as updated driving trajectories, and M sets of representative updated driving trajectories of the M road network control points in the road network control point array are obtained. This process utilizes the common features in historical data to optimize the accuracy and reliability of trajectory prediction, ensuring that the predicted trajectory is closer to the actual driving situation.

[0079] The M groups of updated driving trajectories are mapped to the road network control point array to ensure that the trajectories meet the geometric constraints of the real-time road network topology. The updated trajectory data is integrated into the real-time road network through synchronization operations to generate a visualization of the control point trajectory array including M groups of updated driving trajectory diagrams through M road network control points.

[0080] The control point trajectory array provides basic data for subsequent driving trajectory smoothing processing, ensuring that the predicted trajectory not only meets the vehicle's driving intention but also matches the actual road network environment.

[0081] First, the real-time road network topology is used as a boundary constraint for smoothing to ensure that the generated trajectory strictly follows the geometry of the actual road, including road bends, intersections, and lane changes. Subsequently, mathematical smoothing techniques, such as polynomial interpolation, are used to process the control point trajectory array to generate a smooth and continuous trajectory curve. During the processing, the kinematic characteristics of the vehicle, such as speed and acceleration limits, are fully considered to ensure that the generated trajectory is physically feasible. Ultimately, the smoothed trajectory not only meets the vehicle's driving intention, but also matches the actual road network environment, forming a predicted driving topology.

[0082] The predicted driving topology is decomposed to extract K candidate driving trajectories with local road overlap. These candidate driving trajectories may have overlapping or similar paths in some sections, but are independent and complete as a whole, and together constitute the set of predicted driving trajectories.

[0083] This embodiment outputs a predicted driving trajectory through trajectory prediction, which provides important basic data for subsequent communication network optimization, helps to realize dynamic update and optimization of communication links, and indirectly achieves the technical effect of improving the overall communication performance and safety of traffic scenarios.

[0084] A300: Pre-calculate the base station coverage strength for the predicted driving trajectory and output signal coverage distribution characteristics.

[0085] In one embodiment, the base station coverage strength is pre-calculated for the predicted driving trajectory, and signal coverage distribution characteristics are output. Step A300 of the method provided by the present invention includes:

[0086] A310: Decompose the first candidate driving trajectory to obtain a first trajectory point sequence.

[0087] A320: Using the base station coverage radius as a constraint, search for the first trajectory point sequence for a covering base station, and locate the first covering base station sequence.

[0088] A330: Perform signal propagation prediction on the first coverage base station sequence, and determine the base station signal stability based on the prediction results to obtain a first signal coverage strength sequence.

[0089] A340: Similarly, pre-calculate the base station coverage strength for the K alternative driving trajectories and output K signal coverage strength sequences.

[0090] A350: The K signal coverage strength sequences constitute the signal coverage distribution characteristics.

[0091] In one embodiment, signal propagation prediction is performed on the first coverage base station sequence, and base station signal stability is judged according to the prediction result to obtain a first signal coverage strength sequence. Step A330 of the method provided by the present invention includes:

[0092] A331: Perform signal propagation prediction on the first coverage base station sequence to obtain W groups of signal strength characteristics of W first trajectory points in the first trajectory point sequence.

[0093] A332: Perform base station aggregation on the W groups of signal strength characteristics to obtain H initial serving base stations.

[0094] A333: Perform base station signal stability evaluation based on the W groups of signal strength characteristics, and select a first preferred service base station and H-1 first alternative service base stations from the H initial service base stations based on the evaluation results.

[0095] A334: Using the first preferred service base station as a screening condition, filter out W trajectory point signal coverage strengths from the W groups of signal strength features, and the W trajectory point signal coverage strengths constitute the first signal coverage strength sequence.

[0096] Specifically, in this embodiment, the first alternative driving trajectory is decomposed into a series of discrete trajectory points to obtain W first trajectory points, each of the W first trajectory points contains the coordinate information (longitude, latitude) of the vehicle at that location, and these trajectory points are used for subsequent base station signal coverage analysis.

[0097] The base station coverage radius refers to the maximum distance at which the base station can effectively provide communication services. With the base station coverage radius as a constraint, the first trajectory point sequence is searched for covering base stations to locate the first covering base station sequence, and the first covering base station sequence includes a set of covering base stations corresponding to each of the W first trajectory points.

[0098] It should be understood that each base station coverage set is an information set consisting of base stations that cover a specific trajectory point and can provide communication services for the trajectory point. Through this process, the potential service base station of each trajectory point in the communication network can be clarified, providing a basis for subsequent signal propagation prediction and base station signal stability judgment.

[0099] In this embodiment, signal propagation prediction is performed on the first coverage base station sequence. Specifically, for each trajectory point in the first trajectory point sequence, the signal strength of the corresponding coverage base station is predicted. The signal strength (such as RSRP or RSSI) of each base station to the corresponding covered trajectory point is calculated through a signal propagation model (such as the Okumura-Hata model, the COST-231 model or the free space propagation model). Finally, W groups of signal strength features are obtained, each group of features corresponds to the signal strength information of a trajectory point, and these signal strength features reflect the strength of the base station signal at different locations, providing basic data for the subsequent base station signal stability evaluation.

[0100] The signal strength characteristics of each trajectory point are analyzed, and the base stations with higher signal strength are aggregated into a group to screen out H initial service base stations. The signal strength of these H initial service base stations is good at each trajectory point and can provide timely communication services for the vehicle. It should be noted that the screening of the initial service base stations is based on the average value, maximum value or other statistical indicators of the signal strength to ensure that the selected base stations have higher signal quality.

[0101] The signal strength characteristics of each initial service base station are analyzed to evaluate its signal stability. The signal stability evaluation takes into account factors such as signal strength fluctuations, interference levels, and signal fading. Through the evaluation, a base station with the most stable signal is selected from the H initial service base stations as the first preferred service base station, and the remaining H-1 base stations are selected as the first alternative service base stations.

[0102] It should be understood that the first preferred service base station will be used preferentially in subsequent communications with the target vehicle, while the alternative service base station will provide backup support when necessary to ensure the continuity and stability of communications.

[0103] Taking the first preferred service base station as the benchmark, traverse W trajectory points and extract the signal strength value of the base station at each trajectory point. These values ​​are arranged in the order of trajectory points to form the first signal coverage strength sequence. This sequence records in detail the signal coverage of the first preferred service base station on the first alternative driving trajectory, providing key data support for subsequent communication network optimization. Through this process, it is possible to accurately identify areas with weak signal coverage, plan base station switching or optimize base station layout in advance, ensure the continuity and stability of communication, and provide an accurate basis for potential blind spot identification and channel switching request generation.

[0104] By analogy, the base station coverage strength is pre-calculated for the K candidate driving trajectories, and K signal coverage strength sequences are output. The K signal coverage strength sequences constitute the signal coverage distribution feature.

[0105] This embodiment generates detailed signal coverage distribution characteristics by pre-calculating the base station coverage strength of the alternative driving trajectory, accurately identifying areas with weak signal coverage strength, and dynamically selecting the optimal base station to ensure the continuity and stability of communication. At the same time, it provides an accurate basis for subsequent communication network optimization, potential blind spot identification, and channel switching request generation, significantly improving the optimization efficiency and reliability of the communication network.

[0106] A400: Modify the signal coverage distribution characteristics based on scene elements to obtain K potential blind area sequences.

[0107] In one embodiment, the signal coverage distribution characteristics are corrected based on scene elements to obtain K potential blind area sequences. Step A400 of the method provided by the present invention includes:

[0108] A410: Calling the K scene element sequences of the K alternative driving trajectories through the Internet.

[0109] A420: Use the K scene element sequences to correct the K signal coverage strength sequences, and output K signal coverage correction sequences.

[0110] A430: Locate the K potential blind spot sequences on the K alternative driving trajectories according to the K signal coverage correction sequences.

[0111] In this embodiment, the influence of terrain, buildings and vegetation on signal propagation is taken into account, and the signal shadow fading is calculated, so as to more accurately determine the actual signal coverage strength of the K alternative driving trajectories.

[0112] Specifically, K scene element sequences of the K alternative driving trajectories are extracted through open GIS and urban traffic information network. Each scene element sequence includes a terrain sequence, a building sequence and a vegetation sequence of the driving trajectory. Each scene element sequence describes the environmental characteristics along the corresponding alternative driving trajectory, which is used to evaluate the complexity of signal propagation.

[0113] In order to correct the signal coverage strength sequence and consider the impact of terrain, buildings and vegetation on signal propagation, the log-normal shadow fading model is used for calculation. First, the environmental characteristics of each trajectory point are determined by combining the terrain, building and vegetation information in the scene element sequence. Then, the path loss is calculated based on these characteristics, usually using the Okumura-Hata model or the free space loss model. Path loss reflects the attenuation of the signal in free space, while shadow fading takes into account the additional attenuation of the signal by environmental obstacles. Shadow fading is modeled using a log-normal distribution, and its standard deviation is selected between 4dB and 13dB according to the complexity of the environment. The corrected signal coverage strength sequence is obtained by adding the shadow fading value to the original signal strength.

[0114] By analogy, the K scene element sequences are used to correct the K signal coverage strength sequences, so as to obtain K signal coverage correction sequences that accurately reflect the signal coverage conditions in the actual communication environment and provide key data support for subsequent communication network optimization.

[0115] By analyzing the signal coverage correction sequence of each trajectory point, areas where the signal strength is lower than the preset threshold (such as -100dBm) are identified as potential blind spots. These blind spots may be signal attenuation areas caused by terrain obstruction, building obstruction or dense vegetation.

[0116] This embodiment achieves the technical effect of improving the accuracy of signal coverage evaluation, indirectly provides an important basis for subsequent communication network optimization, and significantly improves the reliability and efficiency of the communication network.

[0117] A500: Perform driving characteristic analysis on the K potential blind spot sequences and output K channel switching request sequences.

[0118] In one embodiment, the driving characteristics of the K potential blind spot sequences are analyzed to output K channel switching request sequences. Step A500 of the method provided by the present invention includes:

[0119] A510: Use the first potential blind spot sequence to traverse the W first trajectory points in the first trajectory point sequence to locate L first blind spot points.

[0120] A520: Use the L first blind spot points as coverage position screening conditions, use the H-1 first candidate service base stations as base station screening conditions, and obtain L groups of blind spot signal coverage strengths from the W groups of signal strength characteristics.

[0121] A530: Serialize the L groups of blind spot signal coverage strengths within the group, and select a switching base station based on the sorting results to obtain L base station switching requests.

[0122] A540: Associate and store the L base station switching requests and the L first blind spots to obtain a first channel switching request sequence.

[0123] A550: Similarly, the driving characteristics of the K potential blind spot sequences are analyzed and the K channel switching request sequences are output.

[0124] Specifically, in this embodiment, first, for the potential blind spot sequence of the first candidate driving trajectory, the W trajectory points in the trajectory point sequence are checked one by one. By analyzing the signal coverage strength of each trajectory point, the trajectory points with signal strength lower than a preset threshold (such as -100dBm) are identified, and these points are blind spot points. Finally, L first blind spot points are determined, which are the key positions on the trajectory where signal coverage is insufficient.

[0125] The positions of the L first blind spots are used as screening conditions, combined with the H-1 first candidate service base stations, and the signal coverage strengths related to these blind spots are extracted from the W groups of signal strength features. Each blind spot corresponds to a group of signal coverage strengths, and a total of L groups of blind spot signal coverage strengths are obtained. These data reflect the signal coverage of the candidate base stations at the blind spots.

[0126] Sort the signal coverage strength of the L groups of blind spots, and select the best base station to switch to based on the order of signal strength. From the H-1 candidate service base stations, select the base station with the strongest and most stable signal as the switching target, and generate L base station switching requests. These requests indicate which base station should be switched to in order to optimize communication quality at a specific blind spot.

[0127] The L base station switching requests are associated with the corresponding L first blind spots and stored to form a first channel switching request sequence. The sequence records in detail the switching base station information corresponding to each blind spot on the first alternative driving trajectory, providing specific instructions for dynamic optimization of the communication network.

[0128] Repeat the above steps to analyze the driving characteristics of all K potential blind spot sequences and generate K channel switching request sequences. Each channel switching request sequence corresponds to an alternative driving trajectory, which contains the switching base station information of all blind spot points on the trajectory, providing comprehensive support for the communication optimization of the target vehicle on different driving paths.

[0129] This embodiment achieves the technical effect of not only optimizing the switching strategy of the communication link and reducing communication interruptions caused by signal blind spots, but also improving the overall performance and reliability of the communication network by accurately identifying potential blind spots on alternative driving trajectories and generating corresponding channel switching request sequences.

[0130] A600: Associate and store the K potential blind area sequences and the K channel switching request sequences to obtain a communication network optimization library.

[0131] Specifically, in this embodiment, K potential blind spot sequences are associated with the corresponding K channel switching request sequences and stored to construct a communication network optimization library. This library integrates the blind spot location information and the corresponding base station switching suggestions, providing comprehensive data support for the dynamic adjustment of the communication network. The communication network optimization library supports real-time update and query, and can dynamically adjust the communication link according to the actual driving path of the vehicle to ensure the continuity and stability of communication.

[0132] A700: According to the real-time spatiotemporal data of the target vehicle, the communication link of the target vehicle is dynamically updated by scheduling switching instructions from the communication network optimization library.

[0133] This embodiment dispatches switching instructions from the communication network optimization library according to the real-time spatial position of the target vehicle, and dynamically updates the communication link of the target vehicle. Specifically, the spatial position information of the vehicle (such as longitude, latitude and elevation) is monitored in real time, and the position information is compared with the potential blind spot sequence stored in the communication network optimization library. When the vehicle enters a potential blind spot, the optimal communication base station is dynamically selected for switching according to the corresponding channel switching request sequence in the communication network optimization library. This process ensures that the vehicle can switch to a base station with a stronger and more stable signal in time during driving, thereby optimizing the communication link, reducing the risk of communication interruption, and significantly improving the reliability and efficiency of the communication network.

[0134] This embodiment achieves the technical effects of improving the vehicle positioning accuracy and the accuracy of motion state estimation, optimizing the reliability of driving trajectory prediction and the dynamic switching efficiency of communication links, improving the data analysis efficiency of dynamic switching of communication links in high-speed mobile communication scenarios, enhancing the reliability and coverage performance of the communication network, and effectively reducing the risk of communication interruption.

[0135] Any of the methods or steps described above may be stored as computer instructions or programs in various types of computer memories, and the computer instructions or programs may be recognized by various types of computer processors to implement any of the methods or steps described above.

[0136] Based on the above specific embodiments of the present invention, any improvements and modifications made to the present invention by those skilled in the art without departing from the principles of the present invention shall fall within the patent protection scope of the present invention.

Claims

1. A communication network optimization system based on spatiotemporal data fusion, characterized in that: The system comprises: A state correction execution module is used to output a real-time correction state vector by fusing and processing the real-time three-dimensional position and real-time inertial data of the target vehicle; A driving trajectory prediction module, used to perform driving trajectory fitting in the road network topology according to the real-time three-dimensional position, and output a predicted driving trajectory; A signal coverage analysis module, used to pre-calculate the base station coverage strength for the predicted driving trajectory and output signal coverage distribution characteristics; A signal coverage correction module, used to correct the signal coverage distribution characteristics based on scene elements to obtain K potential blind area sequences; A channel switching analysis module, used to perform driving characteristic analysis on the K potential blind spot sequences and output K channel switching request sequences; A communication network optimization module, used for associating and storing the K potential blind area sequences and the K channel switching request sequences to obtain a communication network optimization library; A communication optimization execution module, used for scheduling switching instructions from the communication network optimization library to dynamically update the communication link of the target vehicle according to the real-time spatiotemporal data of the target vehicle; The signal coverage analysis module is used for: Decomposing the first candidate driving trajectory to obtain a first trajectory point sequence; Taking the base station coverage radius as a constraint, searching for a covering base station for the first trajectory point sequence, and locating a first covering base station sequence, where the first covering base station sequence includes a covering base station set corresponding to each of the W first trajectory points; Signal propagation prediction is performed on the first coverage base station sequence, and base station signal stability is judged according to the prediction result to obtain a first signal coverage strength sequence, which records in detail the signal coverage of the first preferred service base station on the first alternative driving trajectory, providing key data support for subsequent communication network optimization; Similarly, base station coverage strength is pre-calculated for K candidate driving trajectories, and K signal coverage strength sequences are output; The K signal coverage strength sequences constitute the signal coverage distribution feature; The signal coverage correction module is used for: Calling K scene element sequences of K alternative driving trajectories online; Using the K scene element sequences to correct K signal coverage strength sequences, and outputting K signal coverage correction sequences; Positioning the K potential blind spot sequences on the K alternative driving trajectories according to the K signal coverage correction sequences; The channel switching analysis module is used for: Using the first potential blind area sequence to traverse W first trajectory points in the first trajectory point sequence, and locating L first blind area points; The L first blind spot points are used as coverage position screening conditions, H-1 first candidate serving base stations are used as base station screening conditions, and L groups of blind spot signal coverage strengths are obtained from W groups of signal strength characteristics; The L groups of blind area signal coverage strengths are sequenced within the group, and a switching base station is selected according to the sequencing result to obtain L base station switching requests; Associatively storing the L base station switching requests and the L first blind spots to obtain a first channel switching request sequence; Similarly, the driving characteristics of the K potential blind spot sequences are analyzed and the K channel switching request sequences are output.

2. The communication network optimization system based on spatiotemporal data fusion according to claim 1, characterized in that: The state correction execution module is used for: Interact with the vehicle-mounted GPS to obtain the real-time three-dimensional position of the target vehicle; After calibrating the IMU sensor, running the IMU sensor to collect and obtain the real-time inertial data, wherein the real-time inertial data includes real-time acceleration and real-time angular velocity; Reading real-time driving status parameters through the OBD interface, wherein the real-time driving status parameters include real-time speed and real-time driving direction; The real-time three-dimensional position, the real-time inertial data and the real-time driving state parameters are fused and processed using a Kalman filter algorithm to output the real-time corrected state vector.

3. The communication network optimization system based on spatiotemporal data fusion according to claim 2, characterized in that: The driving trajectory prediction module is used for: Matching the spatiotemporal optimization range according to the real-time driving state parameters, wherein the spatiotemporal optimization range refers to the spatial range in which the vehicle may travel in the future, and is used to limit the calculation range of trajectory prediction to improve calculation efficiency and accuracy; Taking the real-time three-dimensional position as a starting point, performing an associated road network call according to the spatiotemporal optimization range, and outputting a real-time road network topology; The real-time corrected state vector is used as a direction constraint, a driving trajectory fitting is performed on the real-time road network topology, and the predicted driving trajectory is output, wherein the predicted driving trajectory includes K alternative driving trajectories.

4. The communication network optimization system based on spatiotemporal data fusion according to claim 3, characterized in that: The driving trajectory prediction module is used for: Matching the control point demand quantity according to the real-time driving state parameters to obtain the control point quantity constraint; Using the real-time corrected state vector as a direction constraint, locating a road network control point array that satisfies the control point quantity constraint in the real-time road network topology; Using the real-time driving state parameters, predicting future trajectories in the road network control point array, and outputting a control point trajectory array; The real-time road network topology is used as a boundary constraint to smooth the driving trajectory of the control point trajectory array to obtain the predicted driving trajectory.

5. The communication network optimization system based on spatiotemporal data fusion according to claim 1, characterized in that: The signal coverage analysis module is used for: Perform signal propagation prediction on the first coverage base station sequence to obtain W groups of signal strength features of W first trajectory points in the first trajectory point sequence; Performing base station aggregation on the W groups of signal strength characteristics to obtain H initial serving base stations; Performing a base station signal stability evaluation according to the W groups of signal strength characteristics, and screening out a first preferred serving base station and H-1 first candidate serving base stations from the H initial serving base stations according to the evaluation result; The first preferred serving base station is used as a screening condition, and W trajectory point signal coverage strengths are obtained from the W groups of signal strength features, and the W trajectory point signal coverage strengths constitute the first signal coverage strength sequence.

6. The communication network optimization system based on spatiotemporal data fusion according to claim 4, characterized in that: The driving trajectory prediction module is used for: Using the real-time driving state parameters to traverse the driving trajectory change table to obtain a trajectory change response scale; Using the road network control point array as a traffic position retrieval condition and the trajectory change response scale as a traffic time retrieval condition, traffic data networking call is performed to obtain M groups of historical driving trajectories of M road network control points in the road network control point array; By performing a trajectory overlap frequency analysis on the M groups of historical driving trajectories, M groups of updated driving trajectories are screened out; The M groups of updated driving trajectory mappings are synchronized to the real-time road network topology according to the road network control point array to obtain the control point trajectory array.

7. A communication network optimization method based on spatiotemporal data fusion, characterized in that: The method comprises: By fusing and processing the real-time three-dimensional position and real-time inertial data of the target vehicle, a real-time correction state vector is output; Fitting a driving trajectory in a road network topology according to the real-time three-dimensional position, and outputting a predicted driving trajectory; Pre-calculating the base station coverage strength for the predicted driving trajectory and outputting signal coverage distribution characteristics; Modify the signal coverage distribution characteristics based on scene elements to obtain K potential blind area sequences; Performing driving characteristic analysis on the K potential blind spot sequences, and outputting K channel switching request sequences; The K potential blind area sequences and the K channel switching request sequences are stored in association to obtain a communication network optimization library; According to the real-time spatiotemporal data of the target vehicle, a switching instruction is dispatched from the communication network optimization library to dynamically update the communication link of the target vehicle; Pre-calculating the base station coverage strength for the predicted driving trajectory and outputting signal coverage distribution characteristics include: Decomposing the first candidate driving trajectory to obtain a first trajectory point sequence; Taking the base station coverage radius as a constraint, searching for a covering base station for the first trajectory point sequence, and locating a first covering base station sequence, where the first covering base station sequence includes a covering base station set corresponding to each of the W first trajectory points; Signal propagation prediction is performed on the first coverage base station sequence, and base station signal stability is judged according to the prediction result to obtain a first signal coverage strength sequence, which records in detail the signal coverage of the first preferred service base station on the first alternative driving trajectory, providing key data support for subsequent communication network optimization; Similarly, base station coverage strength is pre-calculated for K candidate driving trajectories, and K signal coverage strength sequences are output; The K signal coverage strength sequences constitute the signal coverage distribution feature; The signal coverage distribution characteristics are modified based on scene elements to obtain K potential blind area sequences, including: Calling K scene element sequences of K alternative driving trajectories online; Using the K scene element sequences to correct K signal coverage strength sequences, and outputting K signal coverage correction sequences; Positioning the K potential blind spot sequences on the K alternative driving trajectories according to the K signal coverage correction sequences; Perform driving characteristic analysis on the K potential blind spot sequences and output K channel switching request sequences, including: Using the first potential blind area sequence to traverse W first trajectory points in the first trajectory point sequence, and locating L first blind area points; The L first blind spot points are used as coverage position screening conditions, H-1 first candidate serving base stations are used as base station screening conditions, and L groups of blind spot signal coverage strengths are obtained from W groups of signal strength characteristics; The L groups of blind area signal coverage strengths are sequenced within the group, and a switching base station is selected according to the sequencing result to obtain L base station switching requests; Associatively storing the L base station switching requests and the L first blind spots to obtain a first channel switching request sequence; Similarly, the driving characteristics of the K potential blind spot sequences are analyzed and the K channel switching request sequences are output.

Images