A V2X-based data security processing method, system, device, medium and product

Abstract

The present invention discloses a V2X-based data security processing method, system, device, medium, and product. The method is applied to a system comprising an edge device, a trusted institution, and a cloud enclave. The edge device generates a first public key and a first private key, and sends the first public key to the trusted institution. The cloud enclave generates a second public key and a second private key, and sends the second public key to the trusted institution. The trusted institution generates a neural network key, splits the neural network model into a sub-model set, each of which includes at least two neural network sub-models. Based on the neural network key, the first public key, the second public key, and the sub-model set, encrypted neural network sub-model information is determined and sent to the edge device and the cloud enclave, respectively. By encrypting the sub-model with the public key and the secret key and sending it to the edge device and the cloud enclave, unauthorized access or tampering of the model is prevented, reducing latency and bandwidth requirements for edge devices.

Description

Technical Field

[0001] The present invention relates to the field of vehicle-road collaboration and smart transportation, and in particular to a V2X-based data security processing method, system, equipment, medium and product. Background Art

[0002] In recent years, the development of connected and autonomous vehicles has revolutionized the automotive industry. These vehicles are capable of autonomously navigating roads and exchanging real-time data with heterogeneous vehicle devices, potentially providing intelligent transportation systems with enhanced safety, efficiency, and mobility. Autonomous vehicles leverage vehicle-to-everything (V2X) technology, enabling diverse interactive vehicle applications. However, autonomous vehicles alone are insufficient for developing smart cities. Building intelligent roads, equipped with advanced sensors, network components, and cutting-edge technologies for data analysis and dissemination of relevant information to connected vehicle entities, is also crucial to improving road safety.

[0003] Deep Neural Network (DNN) uses a multi-layer artificial neural network to extract complex features such as pedestrian movement and traffic conditions from raw V2X input data. As an in-vehicle application, collective perception service helps share real-time information between different intelligent connected traffic entities. This collaborative sharing model helps road users reduce risks and accidents by allowing drivers and autonomous driving systems to have a deeper understanding of the driving environment. Assisted by the classification model of the Road Side Unit (RSU), it can achieve real-time decision-making, navigation in complex environments, and adaptation to dynamic conditions, while improving accuracy and reliability.

[0004] However, deploying DNN models in distributed environments such as RSUs and cloud platforms and sharing data may pose problems of adversarial attacks and unauthorized access. The computational overhead required to ensure the confidentiality and integrity of DNN models may hinder the real-time performance required for intelligent transportation applications. Summary of the Invention

[0005] The present invention provides a V2X-based data security processing method, system, device, medium and product to achieve secure transmission of vehicle-road cooperative data.

[0006] According to a first aspect of the present invention, a V2X-based data security processing method is provided, which is applied to a V2X-based data security processing system, wherein the V2X-based data security processing system includes an edge device, a trusted institution, and a cloud enclave. The method includes:

[0007] Generate a first public key and a first private key through the edge device, and send the first public key to the trusted institution;

[0008] Generate a second public key and a second private key through the cloud enclave, and send the second public key to the trusted institution;

[0009] Generate a neural network key through the trusted institution, split the neural network model into sub-model sets, and obtain the sub-model sets including at least two neural network sub-models;

[0010] Determine encrypted neural network sub-model information based on the neural network key, the first public key, the second public key, and the sub-model set, and send each neural network sub-model information to the edge device and the cloud enclave, respectively, so that the edge device and the cloud enclave process the V2X data.

[0011] According to a second aspect of the present invention, a V2X-based data security processing system is provided, which is characterized by comprising: an edge device, a trusted institution, and a cloud enclave.

[0012] The edge device is configured to generate a first public key and a first private key, and send the first public key to the trusted institution;

[0013] The cloud enclave is configured to generate a second public key and a second private key, and send the second public key to the trusted authority;

[0014] The trusted institution is configured to generate a neural network key, perform model splitting on the neural network model to obtain a sub-model set, where the sub-model set includes at least two neural network sub-models; determine encrypted neural network sub-model information based on the neural network key, the first public key, the second public key, and the sub-model set, and send each piece of neural network sub-model information to the edge device and the cloud enclave, respectively, so that the edge device and the cloud enclave process the V2X data.

[0015] According to a third aspect of the present invention, there is provided an electronic device, comprising:

[0016] at least one processor; and

[0017] a memory communicatively connected to the at least one processor; wherein,

[0018] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the V2X-based data security processing method described in any embodiment of the present invention.

[0019] According to a fourth aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the V2X-based data security processing method described in any embodiment of the present invention when executed.

[0020] According to a fifth aspect of the present invention, an embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the V2X-based data security processing method of any embodiment of the present invention.

[0021] The technical solution of the embodiment of the present invention is applied to a V2X-based data security processing system. The V2X-based data security processing system includes an edge device, a trusted institution, and a cloud enclave. A first public key and a first private key are generated by the edge device, and the first public key is sent to the trusted institution; a second public key and a second private key are generated by the cloud enclave, and the second public key is sent to the trusted institution; a neural network key is generated by the trusted institution, and a neural network model is split to obtain a sub-model set, where the sub-model set includes at least two neural network sub-models; encrypted neural network sub-model information is determined based on the neural network key, the first public key, the second public key, and the sub-model set, and the information of each neural network sub-model is sent to the edge device and the cloud enclave respectively. By splitting the neural network model into neural network sub-models, and setting the neural network sub-models encrypted by the neural network key in the edge device and cloud enclave respectively, key decisions can be executed locally on the RSU side, thereby reducing latency and bandwidth requirements, enhancing the privacy of the model, and enhancing the integrity, confidentiality and privacy of the information involved in the model, preventing unauthorized access to such sensitive information, and at the same time utilizing the computing resources of the cloud and trusted execution environment to further improve the security and efficiency of the intelligent road infrastructure.

[0022] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0024] Figure 1 This is a flowchart of a V2X-based data security processing method provided in accordance with the first embodiment of the present invention;

[0025] Figure 2 3. This is a partition time comparison diagram of a V2X-based data security processing method provided in accordance with the first embodiment of the present invention;

[0026] Figure 3 This is a time comparison diagram of a V2X-based data security processing method provided in accordance with the first embodiment of the present invention;

[0027] Figure 4 2 is a schematic diagram of a V2X-based data security processing system according to a second embodiment of the present invention;

[0028] Figure 5 A schematic structural diagram of an electronic device implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0029] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0030] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. 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 clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0031] Example 1

[0032] Figure 1 A flowchart of a V2X-based data security processing method is provided for the first embodiment of the present invention. This embodiment is applicable to the transmission and processing of V2X data. The method can be applied to a V2X-based data security processing system for execution. The V2X-based data security processing system includes an edge device, a trusted organization, and a cloud enclave. The V2X-based data security processing system can be implemented in the form of hardware and / or software. The V2X-based data security processing system can be configured in an electronic device. Figure 1As shown, the method includes:

[0033] S110. Generate a first public key and a first private key through the edge device, and send the first public key to a trusted organization.

[0034] In this embodiment, the edge device can be understood as a device that analyzes data, for example, it can be an RSU in this scenario. The first public key can be understood as the public key of the edge device, and the key pair used together with the first private key is used for the edge device to encrypt data decrypted with the first private key. The first private key can be understood as the key used by the edge device for decryption. The Trusted Authority (TA) can be understood as a central entity responsible for generating the system's global security parameters and all entity keys, responsible for managing security certificates, issuing certificates, and maintaining the ITS ecosystem.

[0035] Specifically, the edge device can generate a first public key and a first private key according to the set encryption and decryption method. The first private key is usually locked by a hardware security module (HSM) (or we can also say it is securely encrypted / stored). Among them, HSM is a computer hardware device used to protect and manage keys used by strong authentication systems and provide related cryptographic operations. When the first private key is needed, it can be temporarily unlocked and then locked again through the HSM. The edge device can use the Transport Layer Security (TLS) protocol to send the first public key to a trusted organization.

[0036] An HSM is integrated into the edge device to securely store encryption keys, prevent unauthorized access to such sensitive information, and ensure the integrity of the encryption process. This strong hardware-based security measure adds an additional layer of protection for neural network models, protecting them from potential threats in edge environments.

[0037] For example, the key generation and subsequent security operations such as encryption and decryption can use Elliptic Curve Cryptography (ECC) with a key size of 256 bits, 512 bits, 768 bits, and 1024 bits. The generated first public key can be obtained by S k Indicates that the first private key can be obtained through P kIn addition, the proposed method for locking and unlocking private keys employs 128-bit S-Box AES encryption for secure operation. ECC or Kyber can be combined with S-Box AES for secure operation. The minimum security robustness of this method is 128-bit symmetric security, which can be achieved with ECC-256 and AES-128. If necessary, the security advantage can be enhanced by implementing larger key sizes.

[0038] S120. Generate a second public key and a second private key through the cloud enclave, and send the second public key to a trusted institution.

[0039] In this embodiment, the Cloud Enclave can be understood as a specialized section of the cloud environment that provides enhanced security and isolation for sensitive data and workloads, ensuring that only authorized users and processes can access data and applications within the Cloud Enclave. The second public key can be understood as the Cloud Enclave's public key. This key pair, used together with the second private key, is used by the Cloud Enclave to encrypt data that is then decrypted using the second private key. The second private key can also be understood as the key used by the Cloud Enclave for decryption.

[0040] Specifically, Yunfeed can generate the second public key and the corresponding second private key according to the set encryption and decryption method. The second private key can be used through ES k Indicates that the second public key is sent to a trusted institution through the TLS protocol, and the second public key can be used through EP k express.

[0041] S130. Generate a neural network key through a trusted institution, split the neural network model, and obtain a sub-model set, where the sub-model set includes at least two neural network sub-models.

[0042] In this embodiment, the neural network key can be understood as the key used to encrypt the neural network model. The neural network model can be understood as the model used for data analysis. The sub-model set can be understood as the set of neural network sub-models after splitting. The neural network sub-model can be understood as the model results after splitting according to different levels, which are used to perform part of the functions of the neural network model.

[0043] Specifically, a neural network key can be generated by a trusted institution according to a set encryption and decryption method. The trusted institution can obtain a neural network model for data analysis and split the neural network model according to a set splitting level. The number of splitting levels can be set according to actual needs to obtain a sub-model set including at least two neural network sub-models.

[0044] S140. Determine encrypted neural network sub-model information based on the neural network key, the first public key, the second public key, and the sub-model set, and send each neural network sub-model information to the edge device and the cloud enclave, respectively, so that the edge device and the cloud enclave process the V2X data.

[0045] Specifically, a trusted institution can use a neural network key to encrypt each neural network sub-model in the sub-model set to obtain an encrypted neural network sub-model, and encrypt the neural network key using a first public key and a second public key respectively to obtain two encrypted neural network keys and distribute them to the corresponding edge devices and cloud enclaves, and distribute the encrypted neural network sub-models to the edge devices and cloud enclaves respectively, wherein the neural network sub-model used for the final decision-making level can be sent to the edge device, so that the edge device and the cloud enclave can process the V2X data based on the neural network sub-model information and determine the final decision result.

[0046] The technical solution of the embodiment of the present invention is applied to a V2X-based data security processing system. The V2X-based data security processing system includes an edge device, a trusted institution, and a cloud enclave. A first public key and a first private key are generated by the edge device, and the first public key is sent to the trusted institution; a second public key and a second private key are generated by the cloud enclave, and the second public key is sent to the trusted institution; a neural network key is generated by the trusted institution, and a neural network model is split to obtain a sub-model set, where the sub-model set includes at least two neural network sub-models; encrypted neural network sub-model information is determined based on the neural network key, the first public key, the second public key, and the sub-model set, and the information of each neural network sub-model is sent to the edge device and the cloud enclave respectively. By splitting the neural network model into neural network sub-models, and setting the neural network sub-models encrypted by the neural network key in the edge device and cloud enclave respectively, key decisions can be executed locally on the RSU side, thereby reducing latency and bandwidth requirements, enhancing the privacy of the model, and enhancing the integrity, confidentiality and privacy of the information involved in the model, preventing unauthorized access to such sensitive information, and at the same time utilizing the computing resources of the cloud and trusted execution environment to further improve the security and efficiency of the intelligent road infrastructure.

[0047] Furthermore, based on the above embodiment, the encrypted neural network sub-model information can be determined according to the neural network key, the first public key, the second public key and the sub-model set as follows:

[0048] Extract the first neural network sub-model and the second neural network sub-model from the sub-model set; determine the encrypted first neural network sub-model information to be sent to the edge device based on the neural network key, the first public key and the first neural network sub-model; determine the encrypted second neural network sub-model information to be sent to the cloud enclave based on the neural network key, the second public key and the second neural network sub-model; and use the first neural network sub-model information and the second neural network sub-model information as the neural network sub-model information.

[0049] In this embodiment, the first neural network sub-model information can be understood as the encrypted neural network sub-model used by the edge device for data processing, which may include the encrypted first neural network sub-model and the encrypted neural network key. The second neural network sub-model information can be understood as the encrypted neural network sub-model used by the cloud enclave for data processing, which may include the encrypted second neural network sub-model and the encrypted neural network key.

[0050] Specifically, a trusted institution extracts the first and second neural network sub-models from the sub-model set. The first neural network sub-model is first encrypted using a neural network key to ensure security during transmission. The neural network key is then encrypted using a first public key to ensure security during transmission. The encrypted first neural network sub-model and the encrypted neural network key are sent to the edge device as first neural network sub-model information, allowing the edge device to decrypt the information using a first private key to extract the correct neural network key and first neural network sub-model. The trusted institution then uses the neural network key to encrypt the second neural network sub-model to ensure security during transmission. The neural network key is then encrypted using a second public key to ensure security during transmission. The encrypted second neural network sub-model and the encrypted neural network key are sent to the cloud enclave as second neural network sub-model information, allowing the cloud enclave to decrypt the information using a second private key to extract the correct neural network key and second neural network sub-model.

[0051] The step of determining the encrypted first neural network sub-model information to be sent to the edge device based on the neural network key, the first public key, and the first neural network sub-model can be concretized as follows:

[0052] The first neural network sub-model is encrypted based on the neural network key to obtain an encrypted first encrypted sub-model; the neural network key is encrypted based on the first public key to obtain an encrypted first encrypted network key; the first encrypted sub-model and the first encrypted network key are used as the encrypted first neural network sub-model information sent to the edge device.

[0053] In this embodiment, the first encrypted sub-model can be understood as the encrypted first neural network sub-model. The first encrypted network key can be understood as the neural network key used to unlock the edge device.

[0054] Specifically, the first neural network sub-model can be encrypted based on the neural network key through the trusted institution to obtain the encrypted first encrypted sub-model, and the neural network key can be encrypted based on the first public key of the edge device to obtain the encrypted first encrypted network key, wherein the first encrypted network key can be obtained through Indicates that the first encrypted sub-model and the first encrypted network key are used together as the encrypted first neural network sub-model information sent to the edge device.

[0055] The step of determining the encrypted second neural network sub-model information to be sent to the cloud enclave based on the neural network key, the second public key, and the second neural network sub-model can be specifically implemented as follows:

[0056] The second neural network sub-model is encrypted based on the neural network key to obtain an encrypted second encrypted sub-model; the neural network key is encrypted based on the second public key to obtain an encrypted second encrypted network key; the second encrypted sub-model and the second encrypted network key are used as the encrypted second neural network sub-model information sent to the cloud enclave.

[0057] In this embodiment, the second encrypted sub-model can be understood as the encrypted second neural network sub-model. The second encrypted network key can be understood as the neural network key used to unlock the cloud enclave.

[0058] Specifically, the second neural network sub-model can be encrypted based on the neural network key through the trusted institution to obtain the encrypted second encrypted sub-model, and the neural network key can be encrypted based on the second public key of the cloud enclave to obtain the encrypted second encrypted network key, wherein the second encrypted network key can be obtained through Indicates that the second encrypted sub-model and the second encrypted network key are used together as the encrypted second neural network sub-model information sent to the cloud enclave.

[0059] As a first optional embodiment of this embodiment, after sending the information of each neural network sub-model to the edge device and the cloud enclave respectively, it also includes:

[0060] Obtain the first private key and the V2X data to be input through Yunfeidi; decrypt the second encrypted network key in the neural network sub-model information based on the first private key to obtain the decrypted first decrypted network key; decrypt the second encrypted sub-model in the neural network sub-model information according to the first decrypted network key to obtain the second neural network sub-model; input the V2X data to be input into the second neural network sub-model, obtain the intermediate output result and send it to the edge device.

[0061] In this embodiment, the input V2X data can be understood as the V2X data used for data analysis. The first decryption network key can be understood as the neural network key after decryption by the cloud enclave. The intermediate output result can be understood as the result processed by the second neural network sub-model.

[0062] Specifically, the cloud enclave can obtain the first private key from the storage medium or the secure cloud enclave for managing private keys, and the cloud enclave can obtain the V2X data to be input transmitted by other devices. The cloud enclave can decrypt the second encrypted network key in the neural network sub-model information based on the first private key to obtain the decrypted first decrypted network key, that is, through ES k To decrypt In order to obtain the first decryption network key The second encrypted sub-model in the neural network sub-model information is decrypted using the first decryption network key to obtain the second neural network sub-model. Yunfeidi can input the V2X data to the second neural network sub-model, obtain the intermediate output result, and send it to the edge device.

[0063] Exemplarily, the neural network model is a trajectory prediction model, the V2X data to be input is vehicle trajectory data, the second neural network sub-model can be the feature extraction part in the neural network model, and the intermediate output result is the feature extraction result. This is only used as an example.

[0064] Furthermore, based on the above embodiment, the present invention further includes:

[0065] Obtain the second private key and the intermediate output result through the edge device; decrypt the first encrypted network key in the neural network sub-model information based on the second private key to obtain the decrypted second decrypted network key; decrypt the first encrypted sub-model in the neural network sub-model information according to the second decrypted network key to obtain the first neural network sub-model; input the intermediate output result into the first neural network sub-model to obtain the final output result.

[0066] In this embodiment, the final output result can be understood as the final result of the neural network model analysis. The second decryption network key can be understood as the neural network key after decryption by the edge device.

[0067] Specifically, the edge device can obtain the second private key from the HSM and receive the intermediate output result sent by the cloud enclave. The edge device can decrypt the first encrypted network key in the neural network sub-model information based on the second private key to obtain the decrypted second decrypted network key, that is, through S k To decrypt In order to obtain the second decryption network key The edge device can decrypt the first encrypted sub-model in the neural network sub-model information according to the second decryption network key to obtain the first neural network sub-model, and input the intermediate output result into the first neural network sub-model to obtain the final output result.

[0068] Exemplarily, the second neural network sub-model may be any part of the neural network model other than the feature extraction part, and the final output result is the trajectory prediction result, which is only used as an example.

[0069] For example, in order to more intuitively understand the time it takes to process data after the neural network model is split and partitioned in the present invention, a comparison is made in the form of a bar graph. Figure 2 This is a partition time comparison diagram of a V2X-based data security processing method provided in Example 1 of the present invention. Figure 2 As shown, the horizontal axis represents different partitioning forms, and the vertical axis represents the time spent on processing data. When the neural network model is set entirely in the cloud, it takes 4.92 seconds for the cloud to process all layers in the neural network model; when all are set on the edge device, it takes the least time, but there will be problems with the security and privacy protection of data and neural network models; when half are in the cloud and half are on the edge device, the time spent is greater than 3 seconds and less than 3.5 seconds; when most of it is in the cloud, it also takes a long time; when most of it is on the edge device, it takes 2.67 seconds, which can be set according to actual needs.

[0070] For example, in order to more intuitively understand the time consumed for decryption and data processing in different scenarios, the following are presented in a comparative form: Figure 3 This is a time comparison diagram of a V2X-based data security processing method provided in Example 1 of the present invention. Figure 3As shown, the grid area represents the time consumed by steps such as decryption, and the black area represents the time consumed by data processing. Scenario 1, which does not utilize partitioning or encryption and performs data processing solely on the edge device, takes 0.53 seconds to process data. Scenario 2, which utilizes decryption and performs data processing on the edge device (equipped with an HSM), increases this time to 2.49 seconds. Scenario 3, which integrates decryption and data processing in both the cloud enclave and the edge device, results in a time of 3.21 seconds. Despite the observed overhead between Scenario 1 and Scenario 3, Scenario 3 offers significantly enhanced security compared to previous approaches. While Scenario 3 introduces an additional overhead of less than 1 second compared to Scenario 2, security is significantly increased. In contrast, Scenario 1, while highly efficient, lacks any security measures, making the model vulnerable. While Scenario 2 enhances edge device security through the HSM, the effect is minimal. Scenario 3 (the system setup described in this invention) offers the highest security, with only a slight increase in overhead.

[0071] Example 2

[0072] Figure 4 This is a structural diagram of a V2X-based data security processing system provided in the second embodiment of the present invention. Figure 4 As shown, the system includes: an edge device 41, a trusted institution 42 and a cloud enclave 43.

[0073] The edge device 41 is configured to generate a first public key and a first private key, and send the first public key to the trusted institution;

[0074] The cloud enclave 42 is configured to generate a second public key and a second private key, and send the second public key to the trusted institution;

[0075] The trusted institution 43 is configured to generate a neural network key, perform model splitting on the neural network model, and obtain a sub-model set, wherein the sub-model set includes at least two neural network sub-models; determine encrypted neural network sub-model information based on the neural network key, the first public key, the second public key, and the sub-model set, and send each neural network sub-model information to the edge device and the cloud enclave, respectively, so that the edge device and the cloud enclave process the V2X data.

[0076] The technical solution of the embodiment of the present invention is applied to a V2X-based data security processing system. The V2X-based data security processing system includes an edge device, a trusted institution, and a cloud enclave. A first public key and a first private key are generated by the edge device, and the first public key is sent to the trusted institution; a second public key and a second private key are generated by the cloud enclave, and the second public key is sent to the trusted institution; a neural network key is generated by the trusted institution, and a neural network model is split to obtain a sub-model set, where the sub-model set includes at least two neural network sub-models; encrypted neural network sub-model information is determined based on the neural network key, the first public key, the second public key, and the sub-model set, and the information of each neural network sub-model is sent to the edge device and the cloud enclave respectively. By splitting the neural network model into neural network sub-models, and setting the neural network sub-models encrypted by the neural network key in the edge device and cloud enclave respectively, key decisions can be executed locally on the RSU side, thereby reducing latency and bandwidth requirements, enhancing the privacy of the model, and enhancing the integrity, confidentiality and privacy of the information involved in the model, preventing unauthorized access to such sensitive information, and at the same time utilizing the computing resources of the cloud and trusted execution environment to further improve the security and efficiency of the intelligent road infrastructure.

[0077] Furthermore, the trusted institution 43 includes:

[0078] A model extraction module, configured to extract the first neural network sub-model and the second neural network sub-model from the sub-model set;

[0079] A first determination module is configured to determine, based on the neural network key, the first public key, and the first neural network sub-model, the encrypted first neural network sub-model information to be sent to the edge device;

[0080] a second determination module for determining, based on the neural network key, the second public key, and the second neural network sub-model, encrypted second neural network sub-model information to be sent to the cloud enclave;

[0081] The third determination module is used to use the first neural network sub-model information and the second neural network sub-model information as neural network sub-model information.

[0082] Furthermore, the first determining module is specifically configured to:

[0083] Encrypting the first neural network sub-model based on the neural network key to obtain an encrypted first encrypted sub-model;

[0084] Encrypting the neural network key based on the first public key to obtain an encrypted first encrypted network key;

[0085] The first encrypted sub-model and the first encrypted network key are used as the encrypted first neural network sub-model information sent to the edge device.

[0086] Furthermore, the second determining module is specifically configured to:

[0087] Encrypting the second neural network sub-model based on the neural network key to obtain an encrypted second encrypted sub-model;

[0088] Encrypting the neural network key based on the second public key to obtain an encrypted second encrypted network key;

[0089] The second encrypted sub-model and the second encrypted network key are sent as the encrypted second neural network sub-model information to the cloud enclave.

[0090] Optionally, the cloud enclave 42 is further configured to: after the trusted institution sends the information of each neural network sub-model to the edge device and the cloud enclave respectively, obtain the first private key and the V2X data to be input;

[0091] Decrypting the second encrypted network key in the neural network sub-model information based on the first private key to obtain a decrypted first decrypted network key;

[0092] Decrypting the second encrypted sub-model in the neural network sub-model information according to the first decryption network key to obtain a second neural network sub-model;

[0093] The V2X data to be input is input into the second neural network sub-model, and an intermediate output result is obtained and sent to the edge device.

[0094] Optionally, the edge device 41 is further configured to:

[0095] Obtaining the second private key and the intermediate output result;

[0096] Decrypting the first encrypted network key in the neural network sub-model information based on the second private key to obtain a decrypted second decrypted network key;

[0097] Decrypting the first encrypted sub-model in the neural network sub-model information according to the second decryption network key to obtain a first neural network sub-model;

[0098] The intermediate output result is input into the first neural network sub-model to obtain the final output result.

[0099] The V2X-based data security processing system provided in the embodiments of the present invention can execute the V2X-based data security processing method according to any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0100] Example 3

[0101] Figure 5 A schematic diagram of the structure of an electronic device 50 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0102] like Figure 5 As shown, the electronic device 50 includes at least one processor 51 and a memory, such as a read-only memory (ROM) 52, a random access memory (RAM) 53, etc., which is communicatively connected to the at least one processor 51. The memory stores a computer program that can be executed by the at least one processor. The processor 51 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 52 or the computer program loaded from the storage unit 58 into the random access memory (RAM) 53. Various programs and data required for the operation of the electronic device 50 can also be stored in the RAM 53. The processor 51, ROM 52, and RAM 53 are connected to each other via a bus 54. An input / output (I / O) interface 55 is also connected to the bus 54.

[0103] Multiple components in the electronic device 50 are connected to the I / O interface 55, including an input unit 56, such as a keyboard, a mouse, etc.; an output unit 57, such as various types of displays, speakers, etc.; a storage unit 58, such as a magnetic disk, an optical disk, etc.; and a communication unit 59, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 59 allows the electronic device 50 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0104] The processor 51 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Examples of the processor 51 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any other suitable processor, controller, or microcontroller. The processor 51 executes the various methods and processes described above, such as the V2X-based data security processing method.

[0105] In some embodiments, the V2X-based data security processing method can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 58. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 50 via ROM 52 and / or communication unit 59. When the computer program is loaded into RAM 53 and executed by processor 51, one or more steps of the V2X-based data security processing method described above can be performed. Alternatively, in other embodiments, processor 51 can be configured to execute the V2X-based data security processing method via any other appropriate means (e.g., via firmware).

[0106] Various embodiments of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0107] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the computer program is executed by the processor, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The computer program may be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0108] In the context of the present invention, computer-readable storage media can be tangible media that can contain or store a computer program for use with an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. Computer-readable storage media can include but are not limited to electronic, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, computer-readable storage media can be machine-readable signal media. More specific examples of machine-readable storage media can include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0109] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0110] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0111] A computing system may include clients and servers. The clients and servers are typically remote from each other and typically interact via a communication network. This client-server relationship arises through computer programs running on the respective computers, creating a client-server relationship. The server may be a cloud server, also known as a cloud computing server or cloud host. This server is a hosting product within the cloud computing service ecosystem that addresses the management difficulties and limited scalability of traditional physical hosting and VPS services.

[0112] In one embodiment, the present invention further includes a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the X-based V2X data security processing method of any embodiment of the present invention.

[0113] The computer program product may be implemented by writing computer program code for performing the operations of the present invention in one or more programming languages, or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0114] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in the present invention can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved. This is not limited herein.

[0115] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

Claims

1. A V2X-based data security processing method, characterized in that: Applied to a V2X-based data security processing system, the V2X-based data security processing system includes an edge device, a trusted institution, and a cloud enclave, the method comprising: Generate a first public key and a first private key through the edge device, and send the first public key to the trusted institution; Generate a second public key and a second private key through the cloud enclave, and send the second public key to the trusted institution; Generate a neural network key through the trusted institution, split the neural network model into sub-model sets, and obtain the sub-model sets including at least two neural network sub-models; Determine encrypted neural network sub-model information based on the neural network key, the first public key, the second public key, and the sub-model set, and send each neural network sub-model information to the edge device and the cloud enclave, respectively, so that the edge device and the cloud enclave process the V2X data.

2. The method according to claim 1, characterized in that Determining encrypted neural network sub-model information according to the neural network key, the first public key, the second public key, and the sub-model set includes: Extracting a first neural network sub-model and a second neural network sub-model from the sub-model set; Determining, based on the neural network key, the first public key, and the first neural network sub-model, encrypted first neural network sub-model information to be sent to the edge device; determining, based on the neural network key, the second public key, and the second neural network sub-model, encrypted second neural network sub-model information to be sent to the cloud enclave; The first neural network sub-model information and the second neural network sub-model information are used as neural network sub-model information.

3. The method according to claim 2, characterized in that Determining, based on the neural network key, the first public key, and the first neural network sub-model, encrypted first neural network sub-model information to be sent to the edge device includes: Encrypting the first neural network sub-model based on the neural network key to obtain an encrypted first encrypted sub-model; Encrypting the neural network key based on the first public key to obtain an encrypted first encrypted network key; The first encrypted sub-model and the first encrypted network key are used as the encrypted first neural network sub-model information sent to the edge device.

4. The method according to claim 2, characterized in that Determining, based on the neural network key, the second public key, and the second neural network sub-model, encrypted second neural network sub-model information to be sent to the cloud enclave includes: Encrypting the second neural network sub-model based on the neural network key to obtain an encrypted second encrypted sub-model; Encrypting the neural network key based on the second public key to obtain an encrypted second encrypted network key; The second encrypted sub-model and the second encrypted network key are sent as the encrypted second neural network sub-model information to the cloud enclave.

5. The method according to claim 1, wherein After sending the neural network sub-model information to the edge device and the cloud enclave respectively, the method further includes: Obtaining the first private key and the V2X data to be input through the cloud enclave; Decrypting the second encrypted network key in the neural network sub-model information based on the first private key to obtain a decrypted first decrypted network key; Decrypting the second encrypted sub-model in the neural network sub-model information according to the first decryption network key to obtain a second neural network sub-model; The V2X data to be input is input into the second neural network sub-model, and an intermediate output result is obtained and sent to the edge device.

6. The method according to claim 5, characterized in that Also includes: Obtaining the second private key and the intermediate output result through the edge device; Decrypting the first encrypted network key in the neural network sub-model information based on the second private key to obtain a decrypted second decrypted network key; Decrypting the first encrypted sub-model in the neural network sub-model information according to the second decryption network key to obtain a first neural network sub-model; The intermediate output result is input into the first neural network sub-model to obtain the final output result.

7. A V2X-based data security processing system, characterized in that: include: Edge devices, trusted institutions, and cloud enclaves, The edge device is configured to generate a first public key and a first private key, and send the first public key to the trusted institution; The cloud enclave is configured to generate a second public key and a second private key, and send the second public key to the trusted authority; The trusted institution is configured to generate a neural network key, perform model splitting on the neural network model to obtain a sub-model set, where the sub-model set includes at least two neural network sub-models; determine encrypted neural network sub-model information based on the neural network key, the first public key, the second public key, and the sub-model set, and send each piece of neural network sub-model information to the edge device and the cloud enclave, respectively, so that the edge device and the cloud enclave process the V2X data.

8. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the V2X-based data security processing method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the V2X-based data security processing method according to any one of claims 1 to 6 when executed.

10. A computer program product, characterized in that The computer program product includes a computer program, which, when executed by a processor, implements the V2X-based data security processing method according to any one of claims 1 to 6.

Images