Intelligent transportation safety monitoring system and method based on quantum encryption and edge computing

Abstract

The invention belongs to the technical field of Internet of Things, and particularly relates to an intelligent transportation safety monitoring system and method based on quantum encryption and edge computing, the system comprises a transportation equipment end and a monitoring center, the transportation equipment end is installed on a transportation tool, and the transportation equipment end comprises a quantum encryption unit, an edge computing node and multiple sensors. The multiple sensors acquire multi-source sensor data, the edge computing device is used for performing format conversion, multi-source data fusion and analysis decision on the multi-source sensor data, and the edge computing device performs data interaction with the monitoring center after performing quantum encryption on data needing to be transmitted through the quantum encryption unit. The monitoring center comprises a quantum key management system, a data storage server and a data analysis server, and the quantum key management system comprises a quantum key generator, a quantum encryption unit and a quantum decryption unit. The method better adapts to the resource limitation of the edge computing device, and meets the requirement of intelligent transportation for real-time performance.

Description

Technical Field

[0001] The present invention belongs to the technical field of Internet of Things, and specifically relates to an intelligent transportation safety monitoring system and method based on quantum encryption and edge computing. Background Art

[0002] With the booming development of smart transportation, the security and processing efficiency of transportation data have become key bottlenecks in the industry's development. Traditional data encryption methods, such as algorithms based on mathematical problems, are no longer able to guarantee absolute security in the face of increasingly powerful computing power and sophisticated cyberattacks. The leakage or tampering of critical data such as cargo information and vehicle travel trajectory during transportation not only results in economic losses but can also lead to serious transportation accidents. Furthermore, with the continuous expansion of transportation scale and the widespread use of IoT devices, the amount of data generated during transportation is exploding. Centralized data processing models that rely on cloud computing centers suffer from high data transmission latency and high network bandwidth pressure, making them unable to meet the stringent real-time and security requirements of smart transportation. Therefore, the challenge is to ensure security while meeting the system's requirements for encryption efficiency and resource consumption, while also avoiding excessive data transmission volumes and overcoming these technical challenges. Summary of the Invention

[0003] The purpose of the present invention is to provide an intelligent transportation safety monitoring system based on quantum encryption and edge computing, which is used to solve the technical problems in the existing technology that it is difficult to meet the system's requirements for encryption efficiency and resource consumption while ensuring security, and to enhance data security during transmission and increase the amount of transmitted data, resulting in high data transmission delays and high network bandwidth pressure.

[0004] The intelligent transportation safety monitoring system based on quantum encryption and edge computing includes a transportation equipment end and a monitoring center. The transportation equipment end is installed on a transportation vehicle. The transportation equipment end includes a quantum encryption unit, an edge computing node and multiple sensors. The multiple sensors collect and obtain multi-source sensor data. The edge computing device is used to convert the format of the multi-source sensor data, fuse the multi-source data and make analysis decisions. The edge computing device quantum encrypts the data to be transmitted through the quantum encryption unit and then exchanges data with the monitoring center. The monitoring center includes a quantum key management system, a data storage server and a data analysis server. The quantum key management system includes a quantum key generator, a quantum encryption unit and a quantum decryption unit. The data stored in the data storage server includes the decrypted data sent by the transportation equipment end. The data analysis server performs monitoring and analysis based on the data stored in the data storage server.

[0005] Preferably, the edge computing device used in the edge computing node integrates a multi-core general-purpose CPU, a neural network processing unit, and a security chip for data encryption; when the means of transport is a transport vehicle, the edge computing device is an edge computing box, and the edge computing device is installed on a bracket made of shock-absorbing and heat-dissipating materials near the vehicle control unit; when the means of transport is a large cargo ship, the edge computing device is an industrial-grade edge computing server deployed on the large cargo ship.

[0006] Preferably, when the goods are stored in a closed container, a small edge computing device powered by wireless energy transmission is set in the container. The small edge computing device is installed inside the container and close to the cargo status monitoring sensor. A quantum encryption device is installed outside the container as a quantum encryption unit.

[0007] The present invention also provides an intelligent transportation safety monitoring method based on quantum encryption and edge computing, which uses the intelligent transportation safety monitoring system based on quantum encryption and edge computing as described above, including the following steps:

[0008] S1, edge computing nodes perform data fusion and analysis based on multi-source sensor data collected by multiple sensors;

[0009] S2. The quantum encryption unit performs quantum encryption on the data to be transmitted and performs data exchange between the transportation equipment and the monitoring center;

[0010] S3. The monitoring center receives data from the transportation equipment and performs quantum decryption, and then stores and analyzes the decrypted data.

[0011] Preferably, the step S1 includes:

[0012] S1.1. Use a combination of rules and machine learning algorithms to perform fusion data format conversion on multi-source sensor data.

[0013] S1.2. Use Kalman filter algorithm to perform multi-source sensor data fusion;

[0014] S1.3. Use genetic algorithms to achieve decision optimization and form a collaborative decision-making mechanism. Based on the collaborative decision-making mechanism, use multiple neural network algorithms to perform real-time analysis on different types of fused data.

[0015] Preferably, in step S1.2, the data collected by the vehicle driving state sensor, the cargo state sensor and the road image sensor are used as input, and these data are fused and processed according to the formula of the Kalman filter algorithm to obtain vehicle, cargo and road state information respectively; in step S1.3, after decision optimization by the genetic algorithm, the convolutional neural network is determined to be used to analyze the road image data and identify road condition information; the long short-term memory network is used to perform in-depth analysis of the vehicle driving data and predict potential vehicle failures; and the isolation forest algorithm is used to detect outliers in the cargo state data.

[0016] Preferably, in step S2, a strategy of multiple distributions and intersection combined with error correction code technology is adopted to generate and distribute mechanical energy keys, and the interval time between each two distributions is adjusted according to the number and frequency of communications; each time the key is distributed, a hash value is generated by a hash algorithm, and the receiving end determines the intersection part as the final key by comparing the hash values; if an intersection judgment error or conflict occurs, Reed-Solomon code is used for processing.

[0017] Preferably, in step S2, block encryption combined with blockchain-based key management technology is used; the image is divided into several small blocks according to the resolution and content characteristics of the video or image, and then an independent quantum key is assigned to each small block for encryption; the generation, distribution and use information of each small block key is recorded on the blockchain, and the consistency and security of the information are ensured through the consensus mechanism of the blockchain; when decryption is required, the system verifies the legitimacy and integrity of the key through the blockchain.

[0018] Preferably, in step S2, when the key is updated, the new key is immediately used to encrypt newly generated data, and the old key is retained for a transition period to decrypt unprocessed old data. During the update process, redundant storage technology is used for data backup, and hash checksum and technology are used for data verification. When decrypting data, the calculated hash value is compared with the stored hash value to determine whether the data has been tampered with.

[0019] Preferably, in step S2, the transportation business is divided into different levels of security based on the sensitivity, importance and scope of influence of the data, and the encryption strategy and key update cycle are adjusted according to the security level; specifically, data involving the core secrets of the transportation enterprise, customer privacy and important cargo information are divided into high-security level business, and a higher level of quantum encryption strategy and a more frequent key update mechanism are adopted; general transportation data are divided into ordinary business, and a relatively low level of encryption strategy and a longer key update cycle are adopted.

[0020] The application has the advantages that the application encodes decision behaviors such as issuing an alarm, adjusting a driving speed, planning a new transportation route and the like into individuals in a genetic algorithm, and finds an optimal decision scheme through operations such as selection, crossover and mutation. The feasibility of replacing an existing convolutional neural network (CNN) with different neural network architectures (such as ResNet and DenseNet) for data processing is explored. After optimization of the selection of different algorithms, the accuracy is ensured, and the resource limitations of edge computing devices are better adapted to, the real-time requirements of intelligent transportation are met. Through an algorithm error processing mechanism, the number of system failures caused by algorithm errors can be reduced by more than 70%. In long-distance freight transportation, problems in algorithm operation can be found and solved in a timely manner, and the continuous and stable operation of the transportation safety monitoring system is ensured.

[0021] Meanwhile, the innovative key generation and distribution strategy of the method improves the key accuracy by about 15%, reaching 95%. This improvement greatly enhances the reliability of the key and effectively reduces the risk of data leakage caused by key errors. Compared with traditional single key distribution, the key accuracy is improved more significantly in complex network environments (such as urban complex networks), ensuring the secure transmission of data in complex networks. The method also combines block encryption with blockchain technology, and small key blocks are transmitted through quantum key distribution technology. In terms of key management, blockchain technology is introduced. By utilizing the decentralized and tamper-proof characteristics of blockchain, a special small key management ledger is constructed. Through the consensus mechanism of the blockchain, the consistency and security of the information are ensured. When decryption is needed, the system verifies the legality and integrity of the key through the blockchain. BRIEF DESCRIPTION OF DRAWINGS

[0022] Figure 1 The figure is a system architecture diagram of the intelligent transportation safety monitoring system based on quantum encryption and edge computing in the application.

[0023] Figure 2 The figure is a quantum encryption process diagram of the intelligent transportation safety monitoring method based on quantum encryption and edge computing in the application.

[0024] Figure 3 The figure is an edge computing data processing flowchart of the intelligent transportation safety monitoring method based on quantum encryption and edge computing in the application. DETAILED DESCRIPTION

[0025] The specific embodiments of the application will be further described in detail below with reference to the accompanying drawings, and the description of the embodiments will help those skilled in the art to have a more complete, accurate and in-depth understanding of the inventive concept and technical solutions of the application.

[0026] As Figure 1-Figure 3As shown, the application provides an intelligent transportation safety monitoring system based on quantum encryption and edge computing, which includes a transportation device end and a monitoring center. The transportation device end is installed on a transportation tool. The transportation device end includes a quantum encryption unit, an edge computing node, and multiple sensors. The multiple sensors collect multi-source sensor data. The edge computing device is used for format conversion, multi-source data fusion, and analysis and decision of the multi-source sensor data. The edge computing device interacts with the monitoring center through quantum encryption of the data to be transmitted. The monitoring center includes a quantum key management system, a data storage server, and a data analysis server. The quantum key management system includes a quantum key generator, a quantum encryption unit, and a quantum decryption unit. The data storage server stores the decrypted data sent by the transportation device end. The data analysis server performs monitoring analysis based on the data stored in the data storage server. The edge computing device used by the edge computing node can use ARM architecture or x86 architecture.

[0027] The edge computing device used by the edge computing node integrates a multi-core general-purpose CPU, an NPU (neural network processing unit) specially designed for deep learning acceleration, and a security chip for data encryption. The device is equipped with a high-speed Ethernet interface to ensure the stability of data transmission, supports USB3.0 and above standards for easy and fast connection of external devices, and has a CANFD interface to adapt to complex sensor networks in intelligent transportation supervision.

[0028] According to the characteristics and needs of different transportation tools, the appropriate edge computing device is selected. On a small delivery truck, a small, low-power edge computing box with real-time data processing capability is installed to meet the requirements of flexible delivery and low power consumption. On a large truck, an industrial-grade edge computing server with strong computing power is deployed to handle large amounts of device data and complex computing tasks. At the same time, according to the device manual, various sensors are accurately installed at key positions of the transportation device, such as installing a gyroscope and an acceleration sensor at a specific position of the vehicle chassis to ensure the accuracy and reliability of data collection.

[0029] High-performance quantum encryption communication terminals are installed on the transportation device and the monitoring center. For long-distance transportation lines, if there is a quantum communication network coverage, directly access the quantum communication network, and fully utilize the high security advantage of quantum encryption; if not covered, use quantum encryption devices combined with traditional encryption devices to ensure the safety of data transmission. For example, in cross-border transportation, quantum encryption communication is used in domestic sections with quantum communication network coverage, and traditional encryption communication is used in foreign regions without quantum communication network coverage. The switching process is automatically completed by an intelligent switching module, which can monitor the network state in real time to ensure seamless connection and security of data transmission.

[0030] Regarding the installation of edge computing equipment on transport vehicles, this solution utilizes novel shock-absorbing and heat-dissipating materials, mounting the edge computing equipment on a specialized bracket near the vehicle's control unit. This mounting method effectively reduces the impact of vibration and high temperatures on the equipment during driving, ensuring stable operation. Quantum encryption equipment is installed in a secure location on the vehicle and shielded to prevent electromagnetic interference, ensuring its proper operation.

[0031] To address the issue of installing edge computing equipment in cargo containers, this solution uses wireless energy transmission technology to power a miniaturized, low-power edge computing device, allowing it to be flexibly installed anywhere inside the container near cargo status monitoring sensors, facilitating real-time data collection. Quantum encryption equipment is installed on the outside of the container, facilitating communication with edge computing equipment on vehicles or large cargo ships, enabling secure data transmission.

[0032] After completing the installation of the transport equipment, including the edge computing device, system integration is required. Data transmission between devices must be carried out using communication protocols such as Ethernet, CAN bus, and RS-485. The electrical characteristics, signal formats, and communication rules of the interfaces must be clearly defined to ensure compatibility and interoperability between devices. The transport equipment's edge computing nodes, quantum encryption equipment, and sensors must be integrated to ensure smooth communication and accurate data exchange between devices. At the monitoring center, the quantum key management system, data storage server, and data analysis server must be integrated to create a unified monitoring and management platform. Through network configuration, secure communication between the transport equipment and the monitoring center is achieved, ensuring safe data transmission and efficient processing.

[0033] Afterwards, comprehensive system testing is carried out, including functional testing, performance testing, security testing, etc.

[0034] (I) Functional testing: simulate various transportation scenarios and abnormal situations to fully verify the system's data collection, encrypted transmission, real-time analysis and decision-making functions. Simulate a vehicle driving at a speed of 130km / h on a highway. The specific steps are: generate vehicle speed data through sensor simulation, input it into the edge computing node, check whether the system can accurately collect speed data, encrypt and transmit it to the monitoring center, identify speed anomalies and issue alarms in time, and plan new reasonable transportation routes at the same time. The expected result is that the system accurately collects speed data within 1 second, the encrypted transmission delay does not exceed 50 milliseconds, an alarm is issued within 2 seconds after identifying the speed anomaly, and a new reasonable transportation route is planned within 5 seconds. Simulate the temperature of the cargo outside the normal range to verify whether the system can monitor temperature changes in time, encrypt and transmit data, trigger alarms and provide corresponding processing suggestions.

[0035] (2) During performance testing: set up a high-load test environment and use the LoadRunner tool to simulate the simultaneous data collection of 100 sensors, processing of high-definition road images (resolution of 1920×1080), etc. The simulated sensor data is generated in the following way: based on the characteristics of different sensors, a random number generator is used in combination with the data distribution patterns in actual transportation scenarios to generate simulated sensor data. Test the system's data processing capabilities and response speed, record indicators such as the time it takes for the system to process a certain amount of data, resource utilization (such as CPU usage, memory occupancy), and evaluate whether the performance meets the requirements of transportation safety monitoring. Under high load conditions, the average time for the system to process 100 sensor data and 10 frames of high-definition images should not exceed 500 milliseconds, the CPU usage should not exceed 80%, and the memory usage should not exceed 70%.

[0036] (3) Security Testing: Using a variety of simulated hacker attack methods, such as quantum attacks (simulating quantum computers' attempts to crack keys) and network intrusions (using common network attack tools and techniques), we test the effectiveness of quantum encryption technology and system security measures. By simulating quantum attacks to crack keys, we verify whether the system's encryption mechanism can withstand attacks and whether data is secure. By simulating network intrusions, we test whether the system's firewalls, intrusion detection systems, and other security measures can detect and block attacks in a timely manner, ensuring that data is not stolen or tampered with. Based on the test results, we conduct targeted optimization and adjustments to the system to ensure stable and reliable operation.

[0037] The present invention also provides an intelligent transportation safety monitoring method based on quantum encryption and edge computing, which includes the following steps.

[0038] S1: The edge computing node performs data fusion and analysis based on multi-source sensor data collected by various sensors. This step specifically includes the following steps.

[0039] S1.1. Use an algorithm that combines rules and machine learning to perform fusion data format conversion on multi-source sensor data.

[0040] For image data, preliminary processing is first performed according to the image data feature extraction rules to extract feature information such as the image's color space and resolution. For images with a resolution higher than the set resolution (such as 1920×1080), bilinear interpolation is used for scaling. For images with a resolution lower than the set resolution, zero padding is used. The calculation formula for bilinear interpolation is:

[0041]

[0042] Among them, (x, y) is the coordinate in the target image, f(x, y) is the pixel value at the coordinate in the target image, (x0, y0), (x1, y0), (x0, y1), (x1, y1) are the coordinates of the four surrounding points in the original image.

[0043] In color space conversion, according to the original color mode, for the common YUV color mode converted to RGB mode, the conversion matrix is:

[0044]

[0045] Then, a convolutional neural network (CNN) is used to learn the features of color conversion and adaptively adjust the conversion parameters.

[0046] The CNN model consists of convolutional layers, pooling layers, and fully connected layers. By training on a large number of image datasets with different color modes and resolutions, the model can automatically adjust the conversion parameters according to the characteristics of the input image and achieve accurate conversion of data formats.

[0047] Testing has shown that for 1,000 images of varying resolutions and color modes, the conversion accuracy of traditional algorithms was 70%, while the proposed algorithm achieved over 90%. Compared to traditional data format conversion algorithms, the proposed algorithm demonstrates a significant advantage when processing low-precision sensor image data. While traditional algorithms can exhibit issues such as color distortion and image blur, causing conversion accuracy to drop below 50%, the proposed algorithm maintains around 80%, effectively improving data processing accuracy and enhancing its adaptability to diverse sensor data types.

[0048] S1.2. Use Kalman filter algorithm to fuse multi-source sensor data.

[0049] This step utilizes the Kalman filter algorithm to fuse multi-source sensor data, improving data accuracy and reliability. The Kalman filter is a recursive filtering algorithm based on linear minimum mean square error estimation. It continuously optimizes the estimate of system status through two steps: prediction and update. In this solution, data collected by vehicle status sensors (such as gyroscopes and accelerometers), cargo status sensors (such as vibration sensors and temperature and humidity sensors), and road image sensors (cameras) are used as input. This data is fused according to the Kalman filter algorithm's formula to obtain more accurate vehicle, cargo, and road status information, respectively.

[0050] S1.3. Use genetic algorithms to achieve decision optimization and form a collaborative decision-making mechanism. Based on the collaborative decision-making mechanism, use multiple neural network algorithms to perform real-time analysis on different types of fused data.

[0051] After multi-source sensor data is fused, the resulting data is extracted separately, including road image data, vehicle driving data, neural network algorithms, and cargo status data. In this step, various neural network algorithms are used to analyze the different data sets in real time. Genetic algorithms, by simulating natural selection and genetic mechanisms, optimize decision-making solutions (i.e., the relationship between different neural network algorithms and different types of fused data). In this solution, the genetic algorithm encodes decision-making behaviors such as issuing an alarm, adjusting driving speed, and planning new transport routes into individuals within the genetic algorithm. Through operations such as selection, crossover, and mutation, the optimal decision solution is found. For example, the feasibility of using different neural network architectures (such as ResNet and DenseNet) to replace existing convolutional neural networks (CNNs) for road image data processing is explored.

[0052] ResNet, by introducing residual connections, effectively addresses the vanishing gradient problem in deep neural networks, potentially offering improved performance when processing complex image data. DenseNet, through dense connections, enhances feature propagation and reuse, theoretically improving model training efficiency and accuracy. However, in actual testing, the high complexity of ResNet and DenseNet models can lead to excessive computational resource consumption when running on edge computing devices, impacting real-time performance. The optimized CNN architecture used in this solution maintains accuracy while better adapting to the resource constraints of edge computing devices, meeting the real-time requirements of intelligent transportation.

[0053] After optimizing decisions using a genetic algorithm, a convolutional neural network (CNN) was selected for analyzing road image data and identifying road conditions. A long short-term memory (LSTM) within a recurrent neural network (RNN) was used to perform in-depth analysis of vehicle driving data and predict potential vehicle failures. An isolation forest algorithm was used to detect outliers in cargo status data. When these analyses detected anomalies, such as abnormal road conditions, potential vehicle failures, or abnormal cargo status, the collaborative decision-making mechanism derived from the optimized decision-making was triggered.

[0054] For step S1, this solution also optimizes the corresponding edge computing algorithm through the following technologies.

[0055] 1) Algorithm error handling: This step establishes an algorithm monitoring mechanism to monitor the accuracy of the algorithm's calculation results, running time and other status in real time, automatically diagnose errors and take corresponding measures.

[0056] If an algorithm error is found during real-time monitoring, the system using this method will automatically perform error diagnosis to determine whether the error is caused by data anomalies (such as data errors caused by sensor failure), model parameter errors (such as parameter deviations during neural network training), or hardware failures (such as calculation errors caused by overheating of the computing chip). Based on the diagnosis results, appropriate measures are taken, such as reinitializing the algorithm, adjusting parameters, or switching to a backup algorithm. At the same time, the error information is recorded in detail, including the time, type, and related data of the error, to facilitate subsequent in-depth analysis and improvement of the algorithm, and continuously improve the stability and reliability of the algorithm. Through the algorithm error handling mechanism, the number of system failures caused by algorithm errors can be reduced by more than 70%. In long-distance freight, problems in the operation of the algorithm can be discovered and resolved in a timely manner to ensure the continuous and stable operation of the transportation safety monitoring system.

[0057] 2) Perform algorithm adaptation and optimization for the hardware platform: Adopt corresponding algorithm optimization strategies for edge computing devices with different architectures.

[0058] The computing power and resource characteristics of different hardware platforms vary significantly, making it difficult for traditional algorithms to fully leverage the advantages of each platform. The edge computing devices used in this solution are either ARM or x86 architectures, and different architectures require different algorithm optimization strategies.

[0059] On ARM-based devices, the MobileNet model was tailored and optimized based on transportation scenario requirements. By analyzing transportation scene image data, the convolutional and fully-connected layers that contribute less to transportation data processing were identified. The last two convolutional layers and one fully-connected layer in the MobileNet model, which contributed less to feature extraction of transportation scene images, were removed to reduce computational effort and memory usage. Simultaneously, model parameters, such as the convolution kernel size and stride, were adjusted to better suit the computing characteristics of the ARM architecture. After optimization, when running the MobileNet model on ARM-based devices, computational effort was reduced by 30% and memory usage by 25%, while maintaining recognition accuracy for transportation scene images above 90%. This enables ARM-based devices to efficiently process transportation scene image data despite limited resources, meeting the low power consumption and real-time processing requirements of small-scale transportation equipment, such as small delivery trucks.

[0060] On x86 architecture devices, multi-threading technology is used to leverage the parallel computing capabilities of multi-core processors. Data processing tasks are rationally divided into multiple subtasks, and corresponding subtasks are assigned to each thread based on the number of processor cores and task type. For example, for image recognition tasks, different areas of the image can be assigned to different threads for processing. Thread synchronization mechanisms such as mutexes and conditional variables ensure data consistency. In actual applications, when processing 100 frames of high-definition images, the use of multi-threading technology reduced the processing time from 10 seconds to 6 seconds, greatly improving the real-time performance of data processing and meeting the performance requirements of intelligent transportation for edge computing devices under different hardware conditions.

[0061] S2. The quantum encryption unit performs quantum encryption on the data to be transmitted and performs data exchange between the transportation equipment and the monitoring center.

[0062] This step specifically applies the following techniques.

[0063] 1) Key generation and distribution strategy: This step uses the strategy of multiple distribution intersection combined with error correction code technology to mechanically generate and distribute keys.

[0064] This step breaks away from the traditional single-shot key distribution model and adopts a multiple-distribution approach, adjusting the interval between distributions based on the volume and frequency of communications. For example, during peak traffic periods, key distribution occurs every 10 minutes; during low-traffic periods, it occurs every 30 minutes. During each distribution, the key is hashed using a hash algorithm (such as SHA-256). The receiving end compares the hash values ​​to determine the final key. Any intersection errors or conflicts are resolved using Reed-Solomon codes.

[0065] The Reed-Solomon code generator polynomial is determined based on parameters such as the bit error rate and data transmission rate of the quantum channel. Assuming that the quantum channel bit error rate is P and the data transmission rate is R, the generator polynomial G(x) is determined as follows: First, the number of error bits t that need to be corrected is determined based on the bit error rate, using the formula: t = [-log2(1-P)]. Then, 2t roots α,α are generated based on the Galois field GF(2m). 2 ,...,α 2t , where m is determined according to the data transmission rate R and the error correction requirement. The larger the m, the stronger the error correction capability, but the higher the computational complexity. In this solution, the specific value is determined according to the requirements, and then the corresponding polynomial G(x) = (x-α)(x-α is generated based on the 2t roots. 2 )...(x-α 2t ), in this way, it can better adapt to the actual situation of quantum channels and improve the accuracy of keys.

[0066] Through extensive experimental comparisons, traditional single-shot key distribution has an accuracy rate of 80%, while this strategy improves key accuracy by approximately 15%, reaching 95%. This improvement significantly enhances key reliability and effectively reduces the risk of data leakage due to key errors. Compared with traditional single-shot key distribution, this strategy achieves a more significant improvement in key accuracy in complex network environments (such as urban networks). While traditional methods may drop below 60%, this strategy maintains accuracy above 90%, ensuring secure data transmission in complex networks.

[0067] 2) Key management technology: block encryption combined with blockchain-based key management technology.

[0068] To address the encryption challenges of unstructured data (such as surveillance videos and images), this solution divides the image into several small blocks based on its resolution and content (for example, a 1920×1080 resolution image can be divided into 128×128 blocks). Each block is then assigned a separate quantum key for encryption. After block encryption, the small block key is transmitted using the aforementioned quantum key distribution technology.

[0069] Blockchain technology is introduced for key management, leveraging its decentralized and tamper-proof nature to build a dedicated small-block key management ledger. The generation, distribution, and usage of each small-block key are recorded on the blockchain, with the blockchain's consensus mechanism ensuring consistency and security. When decryption is required, the system verifies the key's legitimacy and integrity through the blockchain.

[0070] 3) Specific encryption algorithm: The specific encryption algorithm can adopt the quantum one-time pad (QOTP) algorithm or the encryption algorithm based on quantum entanglement.

[0071] Quantum entanglement-based encryption algorithms exploit the properties of quantum entanglement, theoretically enabling more efficient and secure encryption. However, these algorithms place high demands on quantum devices and, with current hardware, are costly to implement. In contrast, the QOTP algorithm offers greater compatibility and cost-effectiveness in this system, meeting the system's requirements for encryption efficiency and resource consumption while ensuring security.

[0072] 4) Dual-key update guarantee mechanism: When updating the key, the new key is immediately used to encrypt newly generated data, and the old key is retained for a transition period (such as 24 hours) to decrypt unprocessed old data; during the update process, the data is backed up and verified to ensure data integrity and availability.

[0073] Traditional key updates are prone to data loss or decryption failures. This solution utilizes a dual-key mechanism, specifically implementing redundant storage for data backup and hash verification. When decrypting data, the calculated hash value is compared with the stored hash value to determine if the data has been tampered with.

[0074] The dual-key mechanism reduces data loss during key updates to less than 2%. For example, after 500 key update tests, the data loss rate with traditional methods was 10%, while this solution reduced it to 2%. This mechanism ensures data continuity, avoids data loss or corruption caused by key updates, and ensures the stable operation of transportation services. In transportation scenarios such as financial escort, where data continuity is extremely important, the dual-key mechanism effectively prevents data interruptions caused by key updates, ensuring a safe and smooth transport process.

[0075] 5) Adjust encryption strategies based on business security levels: This step divides transportation services into different security levels based on the sensitivity, importance, and scope of impact of the data, and adjusts encryption strategies and key update cycles based on the security levels.

[0076] In this embodiment, this step divides transportation services into high-security and normal services. Data involving core confidential information of transportation companies, customer privacy, and important cargo information, such as transportation route planning and cargo value information, is classified as high-security services and utilizes higher-level quantum encryption strategies and more frequent key update mechanisms to ensure high data security. General transportation data, such as daily vehicle operation statistics, is classified as normal services and utilizes relatively lower-level encryption strategies and longer key update cycles to improve system operational efficiency and resource utilization while ensuring data security.

[0077] This business security level classification method not only ensures data security but also optimizes system resource allocation. For example, in the case of high-value cargo transportation, strengthening encryption and shortening key renewal cycles can reduce the risk of data leakage by over 90%. For general transportation data, appropriately extending the key renewal cycle can increase system resource utilization by 30%, effectively balancing security and efficiency.

[0078] S3. The monitoring center receives data from the transportation equipment and performs quantum decryption, and then stores and analyzes the decrypted data.

[0079] The present invention is described above by way of example based on the accompanying drawings. It is obvious that the specific implementation of the present invention is not limited to the above-mentioned method. As long as various non-substantial improvements are made using the inventive concept and technical solution of the present invention, or the inventive concept and technical solution are directly applied to other occasions without improvement, they are all within the scope of protection of the present invention.

Claims

1. An intelligent transportation safety monitoring system based on quantum encryption and edge computing, characterized by: It includes a transportation equipment end and a monitoring center. The transportation equipment end is installed on a transportation vehicle. The transportation equipment end includes a quantum encryption unit, an edge computing node and multiple sensors. The multiple sensors collect and obtain multi-source sensor data. The edge computing device is used to convert the format of the multi-source sensor data, fuse the multi-source data and make analysis decisions. The edge computing device quantum encrypts the data to be transmitted through the quantum encryption unit and then interacts with the monitoring center. The monitoring center includes a quantum key management system, a data storage server and a data analysis server. The quantum key management system includes a quantum key generator, a quantum encryption unit and a quantum decryption unit. The data stored in the data storage server includes the decrypted data sent by the transportation equipment end. The data analysis server performs monitoring and analysis based on the data stored in the data storage server.

2. The intelligent transportation safety monitoring system based on quantum encryption and edge computing according to claim 1 is characterized by: The edge computing device used in the edge computing node integrates a multi-core general-purpose CPU, a neural network processing unit, and a security chip for data encryption; when the means of transport is a transport vehicle, the edge computing device is an edge computing box, and the edge computing device is installed on a bracket made of shock-absorbing and heat-dissipating materials near the vehicle control unit; when the means of transport is a large cargo ship, the edge computing device is an industrial-grade edge computing server deployed on the large cargo ship.

3. The intelligent transportation safety monitoring system based on quantum encryption and edge computing according to claim 1 is characterized by: When the goods are stored in a closed container, a small edge computing device powered by wireless energy transmission is installed in the container. The small edge computing device is installed inside the container and close to the cargo status monitoring sensor. A quantum encryption device is installed outside the container as a quantum encryption unit.

4. An intelligent transportation safety monitoring method based on quantum encryption and edge computing, characterized by: The intelligent transportation safety monitoring system based on quantum encryption and edge computing according to any one of claims 1 to 3 comprises the following steps: S1, edge computing nodes perform data fusion and analysis based on multi-source sensor data collected by multiple sensors; S2. The quantum encryption unit performs quantum encryption on the data to be transmitted and performs data exchange between the transportation equipment and the monitoring center; S3. The monitoring center receives data from the transportation equipment and performs quantum decryption, and then stores and analyzes the decrypted data.

5. The intelligent transportation safety monitoring method based on quantum encryption and edge computing according to claim 4 is characterized by: The step S1 comprises: S1.

1. Use a combination of rules and machine learning algorithms to perform fusion data format conversion on multi-source sensor data. S1.

2. Use Kalman filter algorithm to perform multi-source sensor data fusion; S1.

3. Use genetic algorithms to achieve decision optimization and form a collaborative decision-making mechanism. Based on the collaborative decision-making mechanism, use multiple neural network algorithms to perform real-time analysis on different types of fused data.

6. The intelligent transportation safety monitoring method based on quantum encryption and edge computing according to claim 5 is characterized by: In step S1.2, the data collected by the vehicle driving status sensor, cargo status sensor, and road image sensor are used as input and fused according to the formula of the Kalman filter algorithm to obtain vehicle, cargo, and road status information respectively. In step S1.3, after decision optimization by the genetic algorithm, the convolutional neural network is determined to be used to analyze the road image data and identify road condition information; the long short-term memory network is used to perform in-depth analysis of the vehicle driving data and predict potential vehicle failures; and the isolation forest algorithm is used to detect outliers in the cargo status data.

7. The intelligent transportation safety monitoring method based on quantum encryption and edge computing according to claim 4 is characterized by: In step S2, a strategy of multiple distributions and intersection combined with error correction code technology is adopted to generate and distribute mechanical energy keys, and the interval between each two distributions is adjusted according to the number and frequency of communications; each time the key is distributed, a hash value is generated by a hash algorithm, and the receiving end determines the intersection part as the final key by comparing the hash values; if an intersection judgment error or conflict occurs, the Reed-Solomon code is used for processing.

8. The intelligent transportation safety monitoring method based on quantum encryption and edge computing according to claim 4 is characterized by: In step S2, block encryption is used in combination with blockchain-based key management technology; According to the resolution and content characteristics of the video or image, the image is divided into several small blocks, and then an independent quantum key is assigned to each small block for encryption; the generation, distribution and use information of each small block key is recorded on the blockchain, and the consistency and security of the information are ensured through the blockchain's consensus mechanism; when decryption is required, the system verifies the legitimacy and integrity of the key through the blockchain.

9. The intelligent transportation safety monitoring method based on quantum encryption and edge computing according to claim 4 is characterized in that: In step S2, when the key is updated, the new key is immediately used to encrypt newly generated data. The old key is retained for a transition period to decrypt unprocessed old data. During the update process, redundant storage technology is used for data backup, and hash checksums and other techniques are used for data verification. When decrypting data, the calculated hash value is compared with the stored hash value to determine whether the data has been tampered with.

10. The intelligent transportation safety monitoring method based on quantum encryption and edge computing according to claim 4 is characterized in that: In step S2, transportation services are divided into different security levels based on the sensitivity, importance, and scope of impact of the data, and the encryption strategy and key update cycle are adjusted according to the security level; specifically, data involving core confidential information of transportation companies, customer privacy, and important cargo information are classified as high-security level services, and higher-level quantum encryption strategies and more frequent key update mechanisms are adopted; general transportation data are classified as ordinary services, and relatively lower-level encryption strategies and longer key update cycles are adopted.

Images