Intelligent highway real-time traffic broadcast system and method based on multi-source data fusion

Abstract

The application relates to the technical field of intelligent transportation, and provides a real-time road condition broadcasting system and method for Zhihangtong expressway based on multi-source data fusion. The system comprises five modules, namely, a multi-source data acquisition and fusion module, a broadcasting transmission module, an information processing and publishing module, a system integration and optimization module, and a safety and privacy protection module. The multi-source data acquisition and fusion module collects data through GPS floating cars, video monitoring, traffic flow detectors and the like, and eliminates errors of a single data source by using a space-time data fusion algorithm. The broadcasting transmission module realizes efficient transmission by means of 5G and V2X communication technology. The information processing and publishing module performs real-time processing, prediction and personalized pushing based on big data analysis and artificial intelligence algorithms. The system integration and optimization module optimizes resource allocation by using cloud computing and edge computing. The safety and privacy protection module safeguards data security and user privacy by using blockchain and differential privacy technology. The purpose is to improve the accuracy, real-time performance and personalized service level of road condition information.

Description

Technical Field

[0001] This invention relates to the field of intelligent transportation technology, and in particular to a smart highway real-time traffic broadcasting system and method based on multi-source data fusion. Background Technology

[0002] With the continuous improvement of the highway network and the sustained growth of motor vehicle ownership, real-time traffic information has become crucial for ensuring smooth traffic flow and driver safety. Currently, existing traffic broadcasting technologies have several limitations: data collection sources are singular, relying heavily on single types of sensors, resulting in incomplete traffic information and insufficient accuracy in complex traffic scenarios; broadcasting strategies lack flexibility, making it difficult to dynamically adjust based on the urgency of traffic conditions and data confidence levels, leading to delays in critical information transmission or redundant information consuming resources; information processing and dissemination have low personalization, failing to meet the travel needs of different users; and data security and privacy protection measures are inadequate, posing a risk of information leakage. Therefore, developing a reliable and secure traffic broadcasting system and method that integrates multi-source data, dynamically adjusts broadcasting strategies, provides personalized services, and offers comprehensive coverage is of great significance. Summary of the Invention

[0003] In view of this, the present invention proposes a smart highway real-time traffic broadcasting system and method based on multi-source data fusion, aiming to solve the problems of one-sided information and inflexible broadcasting strategies caused by the single data collection source in the existing highway traffic broadcasting technology.

[0004] The solution provided by the first aspect of the present invention includes:

[0005] A smart highway real-time traffic broadcasting system based on multi-source data fusion includes:

[0006] The multi-source data acquisition and fusion module collects road condition data from multiple data sources and outputs calibrated road condition information using a spatiotemporal data fusion algorithm.

[0007] The broadcast transmission module receives the calibration road condition information, and uses 5G and V2X communication technologies to dynamically adjust the broadcast frequency and content format based on the urgency of the road condition and the network status. Its broadcast strategy is driven by the confidence level and urgency of the calibration road condition information.

[0008] The information processing and publishing module, based on the calibrated road condition information, predicts road condition evolution through a spatiotemporal graph convolutional neural network model, and generates personalized push strategies based on users' historical behavior.

[0009] The system integration and optimization module deploys a hybrid architecture consisting of a cloud computing center and edge computing nodes, connecting the multi-source data acquisition and fusion module, the broadcast transmission module, and the information processing and publishing module, and dynamically allocates computing resources according to the real-time requirements of the task.

[0010] The security and privacy protection module receives the calibrated road condition information, performs differential privacy processing on it, and ensures data integrity through blockchain storage. It also interacts with other modules to achieve end-to-end data security protection.

[0011] Furthermore, in the multi-source data acquisition and fusion module, the multiple data acquisition sources include GPS floating cars, video surveillance, and traffic flow detectors.

[0012] Furthermore, in the multi-source data acquisition and fusion module, the spatiotemporal data fusion algorithm uses a fusion formula... The road condition data collected by GPS floating cars, video surveillance, and traffic flow detectors are fused together;

[0013] in, , , These are GPS floating car data, video surveillance data, and traffic flow detector data, respectively. , , The weights are dynamically adjusted based on different road conditions and data reliability.

[0014] Furthermore, in the broadcast transmission module, the broadcast strategy, driven by both the confidence level and the urgency of the calibration road condition information, specifically includes:

[0015] Through formula Determine broadcast priority ,in, To calibrate the confidence level of road condition information, Depending on the urgency of the road conditions, , These are the weighting coefficients, and ,according to The range of values ​​determines the broadcast strategy.

[0016] Furthermore, when At that time, the broadcast strategy is given the highest priority, 5G slicing technology is used to ensure transmission, the broadcast frequency is no less than once every 3 seconds, and the content format includes high-definition video and text warnings;

[0017] when At that time, the broadcast strategy is medium priority, using V2X communication as the main method and 5G communication as the auxiliary method for transmission. The broadcast frequency is once every 5-10 seconds, and the content format is a combination of text and images.

[0018] when At that time, the broadcast strategy is low priority, conventional 5G communication is used for transmission, the broadcast frequency is more than once every 30 seconds, and the content format is text.

[0019] Furthermore, in the broadcast transmission module, the formula for dynamically adjusting the broadcast frequency is:

[0020]

[0021] in, Depending on the urgency of the road conditions, For bandwidth, For delay, , This is the adjustment coefficient.

[0022] Furthermore, in the information processing and publishing module, the spatiotemporal graph convolutional neural network model is implemented using the formula... The calibrated road condition information is analyzed and predicted;

[0023] in, For input data, This is the weight matrix. For bias vectors, For activation functions;

[0024] The personalized push strategy is expressed through a formula. Generate, where, For user historical behavior data, For preference features, This is a personalized recommendation function.

[0025] Furthermore, the system integration and optimization module uses the formula Allocate computing resources for cloud computing and edge computing, among which, The amount of data processed for edge nodes For task complexity, For real-time requirements, This is the allocation function.

[0026] Furthermore, in the security and privacy protection module, the differential privacy processing is performed using the formula... Processing sensitive user information, including This is the original data. For noise, by controlling the noise variance To protect user privacy, each data block in the blockchain notarization contains the hash value of the previous data block.

[0027] The solution provided by the second aspect of the present invention includes:

[0028] A method for real-time highway traffic broadcasting based on multi-source data fusion includes the following steps:

[0029] Collect road condition data from multiple sources and use a spatiotemporal data fusion algorithm to output calibrated road condition information;

[0030] Upon receiving the calibration road condition information, the broadcast frequency and content format are dynamically adjusted using 5G and V2X communication technologies, combined with the urgency of the road condition and the network status. The broadcast strategy is driven by the confidence and urgency of the calibration road condition information.

[0031] Based on the calibrated road condition information, the road condition evolution is predicted through a spatiotemporal graph convolutional neural network model, and a personalized push strategy is generated based on the user's historical behavior.

[0032] Deploy a hybrid architecture consisting of cloud computing centers and edge computing nodes, and dynamically allocate computing resources according to the real-time requirements of tasks to provide resource support for the above steps;

[0033] Differential privacy processing is performed on the calibrated road condition information, and data integrity is ensured through blockchain notarization, achieving full-process data security protection.

[0034] The technical solution provided in this application has at least the following advantages over the prior art:

[0035] The intelligent highway real-time traffic broadcasting system and method based on multi-source data fusion proposed in this application integrates information from multiple data acquisition sources through a multi-source data acquisition and fusion module and processes it through a spatiotemporal data fusion algorithm. The output calibrated traffic information can comprehensively reflect the actual road conditions, overcoming the problem of one-sided information caused by a single data acquisition source in the prior art. It significantly improves the comprehensiveness and accuracy of traffic information and lays a reliable data foundation for subsequent broadcast transmission, information processing and other links.

[0036] The broadcast transmission module dynamically adjusts the broadcast frequency and content format based on the urgency of road conditions and network status. Furthermore, the broadcast strategy is driven by both the confidence level and urgency of the calibrated road condition information, making the broadcasting method more flexible. This design avoids delays in the transmission of critical information or resource consumption by redundant information, ensuring that important road condition information is delivered to users in a timely and efficient manner, thus improving the effectiveness of information transmission.

[0037] The information processing and dissemination module, based on calibrated traffic information, predicts traffic evolution through a spatiotemporal graph convolutional neural network model and combines this with users' historical behavior to generate personalized push strategies, thus changing the previous situation of low personalization in information processing and dissemination. It can provide customized traffic services for different users' travel needs, greatly improving the user experience and meeting diverse travel information requirements.

[0038] The system integration and optimization module employs a hybrid architecture comprised of a cloud computing center and edge computing nodes. It dynamically allocates computing resources based on the real-time requirements of tasks, achieving rational resource configuration. This makes data processing more efficient, improves the overall system operating efficiency, and ensures smooth collaboration among all modules.

[0039] The security and privacy protection module performs differential privacy processing on the calibrated road condition information and ensures data integrity through blockchain storage. It also interacts with other modules to achieve end-to-end data security protection. This effectively solves the problem of inadequate data security and privacy protection measures in existing technologies, prevents information leakage, ensures data security and reliability, and enhances user trust in the system.

[0040] In summary, this application comprehensively solves many limitations of existing highway traffic broadcasting technology, and significantly improves the accuracy, transmission efficiency, personalized services and security of traffic information, thus having important application value. Attached Figure Description

[0041] Figure 1 This is a structural diagram of the Smart Travel Expressway Real-time Traffic Broadcasting System based on multi-source data fusion, as shown in some embodiments of this application.

[0042] Figure 2 This is an exemplary flowchart of a method for real-time traffic broadcasting on highways based on multi-source data fusion, according to some embodiments of this application; Detailed Implementation

[0043] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0044] This specific embodiment is merely an explanation of this application and is not intended to limit it. Those skilled in the art, after reading this specification, can make modifications to this embodiment without contributing any inventive step, but such modifications are protected by patent law as long as they fall within the scope of the claims of this application. To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, 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, not all, of the embodiments of this application. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of this application.

[0045] The term "comprising" and any variations thereof in the specification and claims of this application are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to such process, method, product or device.

[0046] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design. Specifically, the use of the terms "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0047] The embodiments of this application will now be described in further detail with reference to the accompanying drawings.

[0048] Figure 1 This application presents an exemplary embodiment of a smart highway real-time traffic broadcasting system based on multi-source data fusion, comprising:

[0049] The multi-source data acquisition and fusion module collects road condition data from multiple data sources and outputs calibrated road condition information using a spatiotemporal data fusion algorithm.

[0050] The broadcast transmission module receives the calibration road condition information, and uses 5G and V2X communication technologies to dynamically adjust the broadcast frequency and content format based on the urgency of the road condition and the network status. Its broadcast strategy is driven by the confidence level and urgency of the calibration road condition information.

[0051] The information processing and publishing module, based on the calibrated road condition information, predicts road condition evolution through a spatiotemporal graph convolutional neural network model, and generates personalized push strategies based on users' historical behavior.

[0052] The system integration and optimization module deploys a hybrid architecture consisting of a cloud computing center and edge computing nodes, connecting the multi-source data acquisition and fusion module, the broadcast transmission module, and the information processing and publishing module, and dynamically allocates computing resources according to the real-time requirements of the task.

[0053] The security and privacy protection module receives the calibrated road condition information, performs differential privacy processing on it, and ensures data integrity through blockchain storage. It also interacts with other modules to achieve end-to-end data security protection.

[0054] The intelligent highway real-time traffic broadcasting system and method based on multi-source data fusion proposed in this application integrates information from multiple data acquisition sources through a multi-source data acquisition and fusion module and processes it through a spatiotemporal data fusion algorithm. The output calibrated traffic information can comprehensively reflect the actual road conditions, overcoming the problem of one-sided information caused by a single data acquisition source in the prior art. It significantly improves the comprehensiveness and accuracy of traffic information and lays a reliable data foundation for subsequent broadcast transmission, information processing and other links.

[0055] The broadcast transmission module dynamically adjusts the broadcast frequency and content format based on the urgency of road conditions and network status. Furthermore, the broadcast strategy is driven by both the confidence level and urgency of the calibrated road condition information, making the broadcasting method more flexible. This design avoids delays in the transmission of critical information or resource consumption by redundant information, ensuring that important road condition information is delivered to users in a timely and efficient manner, thus improving the effectiveness of information transmission.

[0056] The information processing and dissemination module, based on calibrated traffic information, predicts traffic evolution through a spatiotemporal graph convolutional neural network model and combines this with users' historical behavior to generate personalized push strategies, thus changing the previous situation of low personalization in information processing and dissemination. It can provide customized traffic services for different users' travel needs, greatly improving the user experience and meeting diverse travel information requirements.

[0057] The system integration and optimization module employs a hybrid architecture comprised of a cloud computing center and edge computing nodes. It dynamically allocates computing resources based on the real-time requirements of tasks, achieving rational resource configuration. This makes data processing more efficient, improves the overall system operating efficiency, and ensures smooth collaboration among all modules.

[0058] The security and privacy protection module performs differential privacy processing on the calibrated road condition information and ensures data integrity through blockchain storage. It also interacts with other modules to achieve end-to-end data security protection. This effectively solves the problem of inadequate data security and privacy protection measures in existing technologies, prevents information leakage, ensures data security and reliability, and enhances user trust in the system.

[0059] In summary, this application comprehensively solves many limitations of existing highway traffic broadcasting technology, and significantly improves the accuracy, transmission efficiency, personalized services and security of traffic information, thus having important application value.

[0060] In some embodiments, the multiple data acquisition sources in the multi-source data acquisition and fusion module include GPS floating cars, video surveillance, and traffic flow detectors.

[0061] Among them, GPS floating cars can provide micro-level individual data such as real-time vehicle location, speed, and driving trajectory; video surveillance can capture macro-level scene information such as overall road traffic flow status, lane occupancy, and sudden accidents; and traffic flow detectors can accurately count quantitative data such as traffic volume, average vehicle speed, and headway. The combination of the three forms a three-dimensional data system of "micro-macro-quantitative," making up for the coverage blind spots of a single data source.

[0062] For example, in the curve area of ​​a highway, GPS floating cars can reflect the speed of vehicles turning the curve, video surveillance can identify whether there are vehicles stuck in the curve, and traffic flow detectors can count the traffic flow changes in the section. When the three are integrated, the complex road conditions in the curve area can be fully presented, avoiding the information loss caused by the limited perspective of a single device.

[0063] In this embodiment, data from different acquisition sources can corroborate each other, effectively filtering outliers or erroneous information. When data from one device deviates, data from other devices can be used as a calibration basis, improving the reliability of the final output calibrated road condition information.

[0064] For example, if a traffic flow detector shows a sudden drop in vehicle speed on a certain road segment, but GPS floating car data and video surveillance both show that vehicles are passing normally, the system can determine that it is a temporary malfunction of the detector. By integrating information from other data sources, the system can eliminate errors and ensure accurate road condition judgment.

[0065] Furthermore, different devices perform differently in extreme environments: GPS signals may weaken in obstructed areas such as tunnels, video surveillance accuracy decreases in heavy rain and fog, and traffic flow detectors are limited by their installation location and cannot cover the entire road segment. Multi-source combinations can leverage the complementary strengths of different devices to ensure stable data collection even in complex scenarios. For example, in heavy rain when video surveillance footage is blurry, speed data from GPS floating vehicles and traffic flow data from traffic flow detectors can serve as primary data sources, ensuring the system's continuous monitoring capability of road conditions.

[0066] In some embodiments, in the multi-source data acquisition and fusion module, the spatiotemporal data fusion algorithm uses a fusion formula. The road condition data collected by GPS floating cars, video surveillance, and traffic flow detectors are fused together; among them, , , These are GPS floating car data, video surveillance data, and traffic flow detector data, respectively. , , The weights are dynamically adjusted based on different road conditions and data reliability.

[0067] In this embodiment, dynamic weights are used to adapt to complex scenarios, improving data accuracy. Different data sources have varying reliability in different scenarios, and dynamic weights can specifically amplify the impact of highly reliable data. For example:

[0068] In a sunny daytime scenario: video surveillance can clearly identify the number of vehicles and lane occupancy, with high data reliability. In this case, [the following settings can be configured]. =0.2, =0.5, =0.3, video surveillance data is given priority, and the fusion result is... It can more accurately reflect the state of road congestion;

[0069] In a rainstorm at night: video surveillance is affected by light and weather, resulting in high data noise, while traffic flow detectors and GPS floating cars are less affected. In this situation, adjustments can be made to... =0.4, =0.1, =0.5, by strengthening the weights of detector and floating car data, to avoid fusion errors caused by video interference.

[0070] In this embodiment, , , The adjustment logic is based on data reliability scores and scenario adaptation rules, and achieves automatic dynamic adjustment through machine learning models. For example:

[0071] Set a data reliability score ( , , The system performs real-time reliability assessments on three types of data sources (GPS floating car data, video surveillance data, and traffic flow detector data), with a rating range of 0-1 (1 indicating complete reliability).

[0072] Affected by GPS signal strength (reduced when obstructed by tunnels) and floating car density (reduced when vehicles are sparse);

[0073] Affected by light intensity (reduced at night), weather transparency (reduced during heavy rain / fog), and image sharpness (reduced when the lens is contaminated);

[0074] Affected by the equipment's self-test status (reduced when there is a fault) and the amplitude of data fluctuations (reduced when there are sudden changes).

[0075] Set scene correction coefficient ( , , The weights are adjusted based on road scene characteristics (such as curves, tunnels, toll stations), for example:

[0076] Tunnel scene: Reduce (weak GPS signal) Improved (more stable detector);

[0077] Toll booth scenario: Improve (ability to identify queue length) Improve (vehicle flow statistics);

[0078] Specifically, , , Automatic adjustment implementation process:

[0079] The system receives raw data from three types of data sources every second and simultaneously collects auxiliary parameters (such as weather, lighting, and equipment status).

[0080] Example: Data collected on a straight road section during a rainstorm at night. =0.7 (GPS signal is stable but there are few vehicles) =0.3 (video image is blurry) =0.8 (detector is operating normally), and the scene correction coefficient is 1 (no special scene).

[0081] The weights are automatically calculated using a weighted normalization algorithm:

[0082]

[0083]

[0084]

[0085] Example: Substituting the above data, we can calculate the following for nighttime rainstorm scenarios: =0.389、 =0.167、 =0.444.

[0086] In addition, the system can display the fusion results every 5 minutes. The error value was calculated by comparing it with the actual road conditions inspected manually. Furthermore, the reliability scoring model is corrected using the gradient descent algorithm (e.g., when the error of video data remains high in light rain, the reliability score for that scenario is automatically reduced). (Base value), enabling self-learning of the weight adjustment strategy.

[0087] In some embodiments, in the broadcast transmission module, the broadcast strategy, driven by both the confidence level and the urgency of the calibration road condition information, specifically includes:

[0088] Through formula Determine broadcast priority ,in, To calibrate the confidence level of road condition information, Depending on the urgency of the road conditions, , These are the weighting coefficients, and ,according to The range of values ​​determines the broadcast strategy.

[0089] Among them, through confidence level and urgency The weighted fusion transforms the vague concept of "importance" into quantifiable priority values. This makes broadcasting strategies more objective. If only based on urgency... This could lead to low-confidence "suspected incidents" being misclassified as high-priority incidents, resulting in wasted resources; relying solely on confidence levels... It may overlook sudden situations with high urgency but low confidence level (such as a recent accident).

[0090] Among them, the weighting coefficient , It can be dynamically adjusted according to road type or time period to adapt to the broadcasting needs of different scenarios, such as:

[0091] Highway main road (large accident impact area): Take =0.7, =0.3, prioritizing the transmission of high-urgency information (such as multi-vehicle rear-end collisions);

[0092] Suburban connecting road (low traffic volume): Take =0.4, =0.6, to avoid low-confidence "small congestion" consuming too many resources.

[0093] In some embodiments, when At this time, the broadcast strategy is given the highest priority, using 5G slicing technology to ensure transmission, with a broadcast frequency of no less than once every 3 seconds, and the content format includes high-definition video and text warnings; for example, in a multi-vehicle rear-end collision during heavy rain ( =0.93), 5G slicing technology can ensure stable transmission of high-definition accident scene video at 30 frames per second, combined with text warnings every 2 seconds, giving drivers within 10 kilometers at least 3 minutes of preparation time to avoid the accident.

[0094] when At this time, the broadcast strategy is of medium priority, primarily using V2X communication to reduce reliance on public networks, while supplementing with 5G communication to fill coverage blind spots. The broadcast frequency is once every 5-10 seconds, and the content format is a combination of text and images; for example, in a slightly congested scenario during the morning rush hour ( =0.575), V2X can directly push congestion images to vehicles within 500 meters, while 5G sends text prompts to vehicles at greater distances, balancing transmission efficiency and resource consumption.

[0095] when At this time, the broadcast strategy is low priority, using conventional 5G communication transmission, with a broadcast frequency of once every 30 seconds or more, and the content format is text. For example, a minor traffic slowdown caused by temporary construction on a certain road section ( =0.4), and a text message "Construction at K50 km, speed 70 km / h" once every 30 seconds is sufficient and will not put extra burden on the network.

[0096] Among them, by broadcast priority By deeply integrating with transmission technology, frequency, and content format, a transmission system with precise "value-resource" matching is formed. This ensures both the "strong penetration" of highly urgent and reliable information and avoids the "resource encroachment" of low-value information, achieving an optimal balance between efficiency, reliability, and economy in highway traffic broadcasting, significantly improving the quality of traffic information services and user satisfaction.

[0097] In some embodiments, the formula for dynamically adjusting the broadcast frequency in the broadcast transmission module is:

[0098]

[0099] in, Depending on the urgency of the road conditions, For bandwidth, For delay, , This is the adjustment coefficient.

[0100] In this embodiment, the urgency of the road condition is... With network status ( bandwidth, The delay is incorporated into a unified calculation framework, avoiding frequency deviations driven by a single factor. When road conditions are urgent (such as accidents or congestion), Increase the push frequency to ensure rapid information transmission; when network bandwidth is sufficient and latency is low... A larger value supports higher frequency broadcasting; however, if the network is congested (low bandwidth, high latency). If the value is low, the frequency will be automatically reduced to avoid network overload.

[0101] In this embodiment, high-frequency broadcasting is triggered only in emergencies and when the network is good, while maintaining a low frequency during normal periods to reduce interference with the driver; at the same time, the frequency is automatically reduced when the network condition is poor to avoid information delays caused by data congestion and to ensure that critical information is "not lost or delayed".

[0102] In some embodiments, in the information processing and publishing module, the spatiotemporal graph convolutional neural network model is implemented using the formula... The calibrated road condition information is analyzed and predicted; wherein, For input data, This is the weight matrix. For bias vectors, The activation function is used; the personalized push strategy is expressed by the formula. Generate, where, For user historical behavior data, For preference features, This is a personalized recommendation function.

[0103] Among them, the spatiotemporal graph convolutional neural network model combines spatial topological relationships with time series features, through the formula... Deeply mining the input multidimensional road condition data can effectively capture the dynamic evolution patterns of road conditions.

[0104] For example: Input data for urban ring expressways during morning rush hour. This includes parameters such as traffic flow, average speed, lane occupancy, and weather conditions (e.g., light rain) for each road segment over the past hour. Weight matrix. The influence of different parameters on road conditions is learned through model training, and the bias vector is used. Used to correct model bias, activation function The ReLU function is used to enhance nonlinear fitting capability. Model output. The system predicts that congestion will occur at a certain hub interchange on the ring road within a certain period of time, providing a reliable predictive basis for subsequent personalized push notifications.

[0105] Among them, through personalized push strategy formula By combining users' historical behavior and preference characteristics, customized information that meets users' needs is generated, avoiding the "one-size-fits-all" approach to information push.

[0106] For example: User A's historical behavior data This indicates that they frequently drive from the east side of the city to the industrial park in the west side of the city between 7:30 and 8:30 am, Monday through Friday, reflecting their preferred driving habits. Prioritizing the shortest route and being sensitive to congestion warnings. Recommendation function. By combining traffic information predicted by a spatiotemporal graph convolutional neural network model, push content is generated. "Monday morning rush hour warning: Your usual shortest route from the east to the west of the city is expected to be congested from 7:50 to 8:20. It is recommended to leave 10 minutes earlier or take an alternative route, which can save about 20 minutes."

[0107] User B's historical behavior data This indicates that they frequently take road trips on weekends, reflecting their preferences. To "Follow service area information along the way", the content will be pushed to you. "Weekend travel tips: There are two service areas along your planned route to Scenic Area A with ample parking spaces. One of the service areas, located 80 kilometers from the starting point, provides charging stations for new energy vehicles."

[0108] Thus, this embodiment, through in-depth analysis of user behavior and preferences, pushes content that is more in line with the actual needs of users, reduces the interference of invalid information on users, and improves users' satisfaction and reliance on traffic information services.

[0109] In some embodiments, the system integration and optimization module uses formulas Allocate computing resources for cloud computing and edge computing, among which, The amount of data processed for edge nodes For task complexity, For real-time requirements, This is the allocation function.

[0110] Among them, the task complexity and real-time requirements As a core parameter, it determines the amount of data processed by the edge nodes. Precisely matched to task characteristics. For tasks with high real-time requirements ( Large value but low complexity ( For tasks with low complexity, edge nodes handle more data processing, reducing latency in data transmission to the cloud; for tasks with high complexity... Large value) but low real-time requirements ( Tasks with low computational value are centrally processed by the cloud computing center, leveraging its powerful computing capabilities. Edge nodes prioritize handling high real-time tasks, avoiding frequent uploads of large amounts of raw data to the cloud and reducing bandwidth consumption and latency in data transmission. For complex tasks with low real-time requirements, centralizing processing in the cloud reduces the computing load on edge nodes and extends device lifespan.

[0111] This embodiment achieves an efficient division of labor by dynamically allocating computing resources, namely, "edge processing of real-time short tasks and cloud processing of non-real-time long tasks". This not only improves the system response speed and resource utilization, but also reduces network load and equipment costs. At the same time, it enhances the system's stability and fault tolerance, providing solid architectural support for the smooth operation of the entire traffic broadcasting system.

[0112] In some embodiments, the differential privacy processing in the security and privacy protection module is performed using a formula. Processing sensitive user information, including This is the original data. For noise, by controlling the noise variance To protect user privacy, each data block in the blockchain notarization contains the hash value of the previous data block.

[0113] Differential privacy processing involves sending data to the original data. Add noise Generate anonymized data And by controlling the noise variance This approach both conceals sensitive user information and preserves the statistical analysis value of the data, preventing the precise identification of individual user information.

[0114] Blockchain stores data through a chain structure, where each data block contains the hash value of the previous data block, forming an immutable distributed ledger. Any modification to historical data will result in a hash mismatch, requiring consensus verification from all nodes in the network, thus technically eliminating the risk of data forgery or tampering.

[0115] Differential privacy processing focuses on "privacy protection during the data release phase," while blockchain notarization focuses on "integrity assurance during the data storage phase." The two complement each other: through the dual mechanism of "dynamic privacy protection + static notarization security," a full-chain data security system is constructed, which not only prevents privacy leaks but also ensures data trustworthiness, providing technical support for the compliant operation of traffic broadcasting systems.

[0116] Figure 1 This application presents an exemplary embodiment of a method for real-time highway traffic broadcasting based on multi-source data fusion, comprising the following steps:

[0117] Step S100: Collect road condition data from multiple sources and output calibration road condition information using a spatiotemporal data fusion algorithm;

[0118] Step S200: Receive the calibration road condition information, and use 5G communication and V2X communication technology to dynamically adjust the broadcast frequency and content format in combination with the urgency of the road condition and the network status. The broadcast strategy is driven by the confidence level and urgency of the calibration road condition information.

[0119] Step S300: Based on the calibrated road condition information, predict the evolution of road conditions through a spatiotemporal graph convolutional neural network model, and generate a personalized push strategy based on the user's historical behavior.

[0120] Step S400: Deploy a hybrid architecture consisting of a cloud computing center and edge computing nodes, and dynamically allocate computing resources according to the real-time requirements of the task to provide resource support for the above steps;

[0121] Step S500: Perform differential privacy processing on the calibrated road condition information and ensure data integrity through blockchain notarization to achieve full-process data security protection.

[0122] The various modules of the aforementioned Smart Travel Expressway Real-Time Traffic Broadcasting System based on multi-source data fusion can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device described in this application can be divided into different functional units or modules to complete all or part of the functions described above.

[0123] An exemplary embodiment of this application provides an electronic device, which may be a server. The electronic device includes a processor, a memory, and a communication interface connected via a system bus. The processor provides computing and control capabilities. The memory can be implemented using any type of volatile or non-volatile storage device or a combination thereof, including but not limited to: disks, optical disks, EEPROMs, EPROMs, SRAMs, ROMs, magnetic storage, flash memory, and PROMs. The memory provides an environment for the operation of the operating system and computer programs stored within it. The communication interface is a network interface used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements the steps of the intelligent highway real-time traffic broadcasting method based on multi-source data fusion described in the above embodiment.

[0124] An exemplary embodiment of this application provides a computer-readable storage medium storing a computer program / instructions that, when executed by a processor, implement the steps of the intelligent highway real-time traffic broadcasting method based on multi-source data fusion described in the various embodiments above. The computer-readable storage medium includes, but is not limited to, ROM, RAM, CD-ROM, magnetic disk, and floppy disk.

[0125] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0126] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and substitutions can be made without departing from the technical principles of the present invention, and these improvements and substitutions should also be considered within the scope of protection of the present invention.

Claims

1. A smart highway real-time traffic broadcasting system based on multi-source data fusion, characterized in that, include: The multi-source data acquisition and fusion module collects road condition data from multiple data sources and outputs calibrated road condition information using a spatiotemporal data fusion algorithm. The broadcast transmission module receives the calibration road condition information, and uses 5G and V2X communication technologies to dynamically adjust the broadcast frequency and content format based on the urgency of the road condition and the network status. Its broadcast strategy is driven by the confidence level and urgency of the calibration road condition information. The information processing and publishing module, based on the calibrated road condition information, predicts road condition evolution through a spatiotemporal graph convolutional neural network model, and generates personalized push strategies based on users' historical behavior. The system integration and optimization module deploys a hybrid architecture consisting of a cloud computing center and edge computing nodes, connecting the multi-source data acquisition and fusion module, the broadcast transmission module, and the information processing and publishing module, and dynamically allocates computing resources according to the real-time requirements of the task. The security and privacy protection module receives the calibrated road condition information, performs differential privacy processing on it, and ensures data integrity through blockchain storage. It also interacts with other modules to achieve full-process data security protection. In the broadcast transmission module, the broadcast strategy is driven by both the confidence level and the urgency of the calibration road condition information, specifically as follows: Through formula Determine broadcast priority ,in, To calibrate the confidence level of road condition information, Depending on the urgency of the road conditions, , These are the weighting coefficients, and ,according to The range of values ​​determines the broadcast strategy; In the information processing and publishing module, the spatiotemporal graph convolutional neural network model is implemented using the formula... The calibrated road condition information is analyzed and predicted; in, For input data, This is the weight matrix. For bias vectors, For activation functions; The personalized push strategy is expressed through a formula. Generate, where, For user historical behavior data, For preference features, This is a personalized recommendation function.

2. The intelligent highway real-time traffic broadcasting system based on multi-source data fusion as described in claim 1, characterized in that: In the multi-source data acquisition and fusion module, the multiple data acquisition sources include GPS floating cars, video surveillance, and traffic flow detectors.

3. The intelligent highway real-time traffic broadcasting system based on multi-source data fusion as described in claim 2, characterized in that: In the multi-source data acquisition and fusion module, the spatiotemporal data fusion algorithm uses a fusion formula. The road condition data collected by GPS floating cars, video surveillance, and traffic flow detectors are fused together; in, , , These are GPS floating car data, video surveillance data, and traffic flow detector data, respectively. , , The weights are dynamically adjusted based on different road conditions and data reliability.

4. The intelligent highway real-time traffic broadcasting system based on multi-source data fusion as described in claim 1, characterized in that: when At that time, the broadcast strategy is given the highest priority, 5G slicing technology is used to ensure transmission, the broadcast frequency is no less than once every 3 seconds, and the content format includes high-definition video and text warnings; when At that time, the broadcast strategy is medium priority, using V2X communication as the main method and 5G communication as the auxiliary method for transmission. The broadcast frequency is once every 5-10 seconds, and the content format is a combination of text and images. when At that time, the broadcast strategy is low priority, conventional 5G communication is used for transmission, the broadcast frequency is more than once every 30 seconds, and the content format is text.

5. The intelligent highway real-time traffic broadcasting system based on multi-source data fusion according to claim 4, characterized in that: The formula for dynamically adjusting the broadcast frequency in the broadcast transmission module is as follows: in, Depending on the urgency of the road conditions, For bandwidth, For delay, , This is the adjustment coefficient.

6. The intelligent highway real-time traffic broadcasting system based on multi-source data fusion according to claim 1, characterized in that: The system integration and optimization module uses the formula Allocate computing resources for cloud computing and edge computing, among which, The amount of data processed for edge nodes For task complexity, For real-time requirements, This is the allocation function.

7. The intelligent highway real-time traffic broadcasting system based on multi-source data fusion according to claim 1, characterized in that: In the security and privacy protection module, the differential privacy processing is performed using the formula... Processing sensitive user information, including This is the original data. For noise, by controlling the noise variance To protect user privacy, each data block in the blockchain notarization contains the hash value of the previous data block.

8. A method for real-time traffic broadcasting on highways based on multi-source data fusion, applied to the real-time traffic broadcasting system for highways based on multi-source data fusion as described in any one of claims 1 to 7, characterized in that, Includes the following steps: Collect road condition data from multiple sources and use a spatiotemporal data fusion algorithm to output calibrated road condition information; Upon receiving the calibration road condition information, the broadcast frequency and content format are dynamically adjusted using 5G and V2X communication technologies, combined with the urgency of the road condition and the network status. The broadcast strategy is driven by the confidence and urgency of the calibration road condition information. Based on the calibrated road condition information, the road condition evolution is predicted through a spatiotemporal graph convolutional neural network model, and a personalized push strategy is generated based on the user's historical behavior. Deploy a hybrid architecture consisting of cloud computing centers and edge computing nodes, and dynamically allocate computing resources according to the real-time requirements of tasks to provide resource support for the above steps; Differential privacy processing is performed on the calibrated road condition information, and data integrity is ensured through blockchain notarization, achieving full-process data security protection.

Images