Internet of vehicles card real-name registration method, device and equipment and storage medium

Abstract

The invention discloses an Internet of Vehicles card real-name registration method, device and equipment and a storage medium, and the method comprises the steps: obtaining a real-name registration request of a user, and judging whether the real-name registration is the first real-name registration or not according to the real-name registration request; if the real-name registration is the first real-name registration, collecting identity information and a short message verification code of the user, and generating a dynamic identity certificate based on the identity information and the short message verification code; encrypting and storing the identity information by taking the dynamic identity credential as a secret key; calling a real-name registration interface to complete real-name registration based on the encrypted identity information; if the real-name registration is not the first real-name registration, acquiring basic information of the user; carrying out risk scoring on the basic information through a dynamic risk assessment model and determining a risk level; and executing corresponding registration processes according to different risk levels. According to the invention, dynamic risk assessment is carried out on the real-name registration behavior of the user through the dynamic risk assessment model, and an identity credential multiplexing mechanism is adopted for the low-risk user, so that the real-name registration efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of vehicle networking technology, and in particular to a method, apparatus, device, and storage medium for real-name registration of vehicle networking cards. Background Technology

[0002] In recent years, vehicle-to-everything (V2X) technology has experienced explosive growth. The deep integration of cutting-edge technologies such as 5G, IoT, and big data has accelerated the transformation of the automotive industry towards intelligence and connectivity. With millions of existing vehicles and a massive influx of new terminals, the application scale of V2X cards continues to expand. However, the security vulnerabilities arising from their anonymized management are becoming increasingly prominent. Malicious users are using unregistered V2X cards to carry out illegal activities such as remote control and data theft, seriously threatening personal privacy and public safety.

[0003] To address this challenge, major telecom operators have successively launched online real-name registration solutions for vehicle networking cards, but many pain points still need to be addressed in practical applications: First, the existing registration process is cumbersome and complex. Users need to submit multiple types of materials, such as ID cards and driver's licenses, while also having their ID card OCR information, on-site photos, and video liveness data collected. For other types of documents besides ID cards, a second manual review is often required. The entire process involves many steps and is time-consuming, resulting in a poor user experience and low registration efficiency.

[0004] Secondly, security risks are prominent. The online real-name registration system faces security threats such as remote attacks and malicious control, and privacy protection and data security issues cannot be ignored. In particular, the use of static identity information during transmission carries the risk of interception, further exacerbating data security vulnerabilities.

[0005] Currently, mainstream solutions for online real-name registration of vehicle-to-everything (V2X) cards have significant limitations in obtaining user identity information: they either rely on a single biometric technology, which is insufficient in security, or they employ complex and cumbersome identity information collection technologies, which are not convenient. These solutions have not yet established an efficient identity credential reuse mechanism, and they lack the ability to dynamically perceive and analyze user behavior risks in real time. They are unable to achieve an effective balance between security and convenience, and thus cannot meet the growing security management needs of the V2X system. Summary of the Invention

[0006] The purpose of this invention is to provide a method, apparatus, device, and storage medium for real-name registration of vehicle network cards, addressing the problems of insufficient security or poor convenience in existing online real-name registration solutions for vehicle network cards.

[0007] In a first aspect, embodiments of the present invention provide a method for real-name registration of a vehicle network card, including: Obtain the user's real-name registration request, and determine whether the real-name registration is the first real-name registration based on the real-name registration request; If the real-name registration is the first real-name registration, the user's identity information and SMS verification code are collected, and a dynamic identity credential is generated based on the identity information and SMS verification code. The identity information is encrypted and stored using the dynamic identity credential as the key; Real-name registration is completed by calling the real-name registration interface based on the encrypted identity information; If the real-name registration is not the first time, then the user's basic information will be obtained; The basic information is scored and its risk level is determined using a dynamic risk assessment model. The corresponding registration process will be carried out according to different risk levels.

[0008] Secondly, embodiments of the present invention provide a vehicle network card real-name registration device, comprising: The judgment unit is used to obtain the user's real-name registration request and determine whether the real-name registration is the first real-name registration based on the real-name registration request; The generation unit is used to collect the user's identity information and SMS verification code and generate a dynamic identity credential based on the identity information and SMS verification code if the real-name registration is the first real-name registration. An encryption unit is used to encrypt and store the identity information using the dynamic identity credential as a key. The registration unit is used to complete real-name registration by calling the real-name registration interface based on encrypted identity information. The acquisition unit is used to acquire the user's basic information if the real-name registration is not the first real-name registration; The scoring unit is used to score the basic information based on a dynamic risk assessment model and determine the risk level. The execution unit is used to perform the corresponding registration process according to different risk levels.

[0009] Thirdly, embodiments of the present invention provide a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the vehicle network card real-name registration method described in the first aspect.

[0010] Fourthly, embodiments of the present invention also provide a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, which, when executed by a processor, implements the vehicle network card real-name registration method described in the first aspect.

[0011] This invention discloses a method, apparatus, device, and storage medium for real-name registration of vehicle network cards. The method includes: obtaining a user's real-name registration request and determining whether the real-name registration is the first time based on the request; if the real-name registration is the first time, collecting the user's identity information and SMS verification code, and generating a dynamic identity credential based on the identity information and SMS verification code; encrypting the identity information using the dynamic identity credential as a key and storing it; calling the real-name registration interface based on the encrypted identity information to complete the real-name registration; if the real-name registration is not the first time, obtaining the user's basic information; performing a risk score on the basic information using a dynamic risk assessment model and determining the risk level; and executing the corresponding registration process according to different risk levels. This invention uses a dynamic risk assessment model to dynamically assess the user's real-name registration behavior, employs an efficient identity credential reuse mechanism for low-risk users, and quickly completes the real-name registration, greatly improving the efficiency of real-name registration while also ensuring security. This invention also provides a vehicle network card real-name registration apparatus, a computer-readable storage medium, and a computer device, which have the above-mentioned beneficial effects, and will not be elaborated further here. Attached Figure Description

[0012] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 A flowchart illustrating the real-name registration process for vehicle network cards; Figure 2 Another flowchart illustrating the real-name registration method for vehicle network cards; Figure 3 A flowchart illustrating the sub-processes of the real-name registration method for vehicle network cards; Figure 4 A schematic block diagram of a vehicle network card real-name registration device; Figure 5 Interactive diagram of the vehicle network card real-name registration system architecture. Detailed Implementation

[0014] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0015] It should be understood that, when used in this specification and the appended claims, the terms “comprising” and “including” indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more of its features, integrals, steps, operations, elements, components and / or collections thereof.

[0016] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0017] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0018] Please see Figure 1 and Figure 2 This embodiment provides a method for real-name registration of vehicle network cards, including: S101: Obtain the user's real-name registration request, and determine whether the real-name registration is the first real-name registration based on the real-name registration request; Specifically, when a user needs to register their vehicle network card online, the system initiates the real-name registration process. The user submits a real-name registration request through a mobile application or web page, which typically includes the user's basic identification information, such as a mobile phone number or a unique device identifier.

[0019] The system receives and parses the real-name registration request, extracting key user identification information. Subsequently, the system uses this key user identification information as a query condition to access the backend blockchain evidence repository or user credential database to search for whether there is a valid dynamic identity credential record associated with the user.

[0020] The core of the judgment logic lies in the query results. If a valid, unexpired dynamic identity credential record exists in the database, the system determines that the user's operation is not the first time registering with their real name. Based on this determination, the system will automatically guide the user to the fast real-name registration channel, prompting the user to submit the vehicle VIN code, vehicle network card ICCID, and existing dynamic identity credential for subsequent credential verification and risk assessment processes.

[0021] Conversely, if the database does not contain any dynamic identity credential records associated with the user's identification information, or if all existing credential records are shown as expired, the system determines that the user is registering for real-name authentication for the first time. Accordingly, the system will guide the user to the standard real-name registration channel, requiring the user to complete a series of necessary information collection steps, including ID card OCR information collection (user's name and ID number), on-site photo taking, liveness detection, and SMS verification code verification, following the standard procedure for first-time registration.

[0022] This judgment mechanism is the foundation for implementing differentiated processing procedures, ensuring that first-time registered users can complete the full information filing to generate credentials, while non-first-time registered users can enjoy fast services using existing credentials, thus balancing security and efficiency overall.

[0023] S102: If the real-name registration is the first real-name registration, then collect the user's identity information and SMS verification code, and generate a dynamic identity credential based on the identity information and SMS verification code; Specifically, collecting user identity information and SMS verification codes, and generating dynamic identity credentials based on these information and codes, includes: Collect user name, ID number, on-site photo, live video, mobile phone number, timestamp, and SMS verification code; Using SMS verification codes as dynamic salts, a hash algorithm is used to concatenate the user's name, ID number, mobile phone number, timestamp, and SMS verification code to obtain dynamic identity credentials.

[0024] More specifically, the system will prompt the user to submit the required identity documents in sequence. The user enters their name and ID number through the client interface. Simultaneously, the system uses the device's camera to guide the user to take a clear photo of the scene and record a short liveness video for authenticity verification. In addition, the user needs to enter their current mobile phone number, and the system will then send a 6-digit SMS verification code to that number. The user then enters this verification code back into the application interface. Throughout the process, the system automatically records the standard timestamp of each operation.

[0025] After all information is collected, the system initiates the dynamic identity credential generation process. The generation process uses the user-submitted SMS verification code as the key dynamic salt value. Employing the MD5 hash algorithm, the user's name, ID number, mobile phone number, system-recorded timestamp, and SMS verification code are concatenated in a predefined order. Subsequently, the MD5 algorithm performs a one-way hash calculation on the concatenated string, ultimately generating a unique, fixed-length 128-bit dynamic identity credential. Dynamic Identity Credential = MD5(ID Number + Name + Mobile Phone Number + Dynamic Verification Code + Timestamp).

[0026] This credential is linked to a timestamp, and its validity period is set to 24 hours from the time of its generation. The system stores this dynamic identity credential, timestamp, SMS verification code, and encrypted user identity information in a blockchain repository for use in subsequent real-name registration or fast-track verification.

[0027] In some embodiments, during the initial real-name registration scenario, the system collects user identity information through the interactive interface of the operator's vehicle networking service APP. Specifically, it calls the user's device camera to take a live photo containing the user's face and records a 10-second live video (requiring the user to blink); it scans the front and back of the user's ID card using the APP's built-in OCR recognition function to extract the user's name and ID number; it prompts the user to manually enter their mobile phone number, and after the user enters the number, the system sends a 6-digit SMS verification code to that mobile phone number. After the user enters the verification code on the interface, the system simultaneously obtains the timestamp of the current operation (accurate to the second).

[0028] The system uses the 6-digit SMS verification code entered by the user as a dynamic salt and processes the collected information using a hash algorithm. During processing, the strings are concatenated in the order of user name, ID number, mobile phone number, timestamp, and SMS verification code. After concatenation, a 128-bit dynamic identity credential is calculated using a hash algorithm.

[0029] In some embodiments, the method further includes: obtaining the International Mobile Equipment Identity (IMEI) code and the current IP address of the user device, reading the IMEI code through the system interface, extracting the IP address from the network connection information, and combining them to generate a device fingerprint identifier; Next, a one-time token (TOTP) synchronized with the device fingerprint identifier and the preset key is generated. The HMAC-SHA1 algorithm (HMAC-SHA1 is a keyed hash algorithm based on the SHA-1 hash function) is used, with the device fingerprint as the key and the timestamp as the input, and the token value is updated every 30 seconds. Then, the verification code sent to the user's mobile phone number via SMS is received, and the verification code and the TOTP value are concatenated in a fixed order to form a dynamic salt value string; The dynamic salt value string is then concatenated with the user's identity information (name, ID number, mobile phone number) and the current timestamp according to a predetermined format. The concatenated result is then encrypted using the SHA-256 hash algorithm to generate a fixed-length dynamic identity credential. Next, the generated dynamic identity credential is verified to meet the preset verification rules, including length verification (e.g., 32 bytes) and format verification (e.g., hexadecimal characters). If the verification passes, the credential, device fingerprint, and TOTP generation time are synchronized and stored in the distributed ledger. If the verification fails, the TOTP is regenerated and the hash concatenation process is repeated. After the credential is generated, the association between the device fingerprint and the TOTP is recorded, and the validity period of the credential is set (e.g., 5 minutes). If the same device fingerprint is detected to be used repeatedly for the same TOTP within the validity period, it is marked as abnormal behavior and a risk warning is triggered.

[0030] This embodiment employs a two-factor authentication mechanism, eliminating the system's complete reliance on SMS verification codes for identity verification. Even in the event of SMS delays or loss, the credential generation process can still be maintained through device fingerprint factors, significantly improving system service availability and user experience.

[0031] S103: Encrypt the identity information using the dynamic identity credential as the key and store it; The system retrieves all previously collected user identity information from the cache. This information includes unstructured image and video data, such as photos of the front and back of an ID card, photos taken of the user on-site, and screenshots of liveness videos, as well as structured text data, such as the user's name, ID number, and mobile phone number.

[0032] Subsequently, the system uses the dynamic identity credential as the encryption key and employs the efficient AES-256 encryption algorithm to encrypt the aforementioned complete identity information dataset. This process transforms sensitive personal information from plaintext into unreadable ciphertext data. The encryption operation ensures that even if the data is intercepted during transmission or storage, attackers cannot decrypt and obtain the original valid information without acquiring the dynamic identity credential.

[0033] After encryption, the system will transmit the generated encrypted data, along with business identifiers related to this real-name registration (such as vehicle VIN code, vehicle network card ICCID, etc.), to the operator's backend real-name registration system via secure HTTPS protocol to complete the core registration business logic.

[0034] Simultaneously, the system generates a record of evidence. This record contains crucial linking information: the generated dynamic identity credential (or its hash value), the timestamp used to generate the credential, the SMS verification code submitted by the user, and an index pointing to the storage location of the encrypted identity information. This record is then sent to a blockchain repository for distributed storage. Leveraging the immutability of the blockchain, the key evidence and connections of this real-name registration operation are permanently secured.

[0035] S104: Complete real-name registration by calling the real-name registration interface based on the encrypted identity information; After the dynamic identity credential is generated and the user's identity information is encrypted, the system immediately enters the real-name registration interface call phase. Essentially, this involves connecting the secure data packet processed by the front-end with the back-end business system to ultimately complete the registration process.

[0036] The system first assembles the encrypted identity information and necessary business data into a request data packet that conforms to the interface specification. The necessary business data includes, but is not limited to, the vehicle's VIN code, the vehicle network card's ICCID number, and a timestamp to identify this operation.

[0037] Subsequently, the system sends the assembled request data packet to the operator's backend real-name registration system interface through a pre-set secure network channel. This interface, acting as a bridge for data interaction between the front-end and back-end, receives and parses this data packet.

[0038] Upon receiving a request, the background real-name registration system does not immediately perform decryption. Its first step is to perform business logic verification, such as confirming whether the vehicle network card is eligible for registration and whether the binding relationship between the VIN code and ICCID is valid. Only after the basic business verification is passed does the system proceed to the decryption stage.

[0039] The decryption process relies on the relationships previously stored in the blockchain's evidence repository. The backend system locates the corresponding evidence record based on the index information in the request packet and retrieves the dynamic identity credential used to generate the encrypted data. The system then uses this dynamic identity credential as a key to decrypt the received ciphertext identity information, restoring the plaintext identity information.

[0040] After decryption and obtaining the plaintext information, the backend system executes the core registration operation, establishing a formal and unique binding relationship between the user's identity information and the vehicle network card information in the database. The entire process of interface call, processing, and registration is fully recorded in the system log. Finally, the backend real-name registration interface returns the operation result, i.e., the registration success or failure status information and the reason, to the frontend system. The frontend system updates the interface status based on the returned result, informing the user of the final completion status of the real-name registration.

[0041] S105: If the real-name registration is not the first real-name registration, then obtain the user's basic information; Once the system determines that a user's real-name registration request is not a first-time registration, it immediately initiates a dedicated process for fast-track real-name registration. The system automatically presents a simplified information input interface to the user, guiding them to submit the basic information necessary to complete the registration.

[0042] On this fast-track interface, users only need to enter three core pieces of basic information: the VIN code of the vehicle to be bound, the ICCID number of the vehicle networking card that needs to be verified with real name, and the dynamic identity credential obtained by the user during the initial registration and saved by the system. This process significantly reduces the amount of data that users need to manually enter.

[0043] S106: The basic information is scored and the risk level is determined by a dynamic risk assessment model; In this embodiment, before performing risk scoring and determining the risk level of basic information using the dynamic risk assessment model, the following steps are included: Verify whether the dynamic identity credential has exceeded the scheduled time; If the dynamic identity credential expires after the scheduled time, the user will be prompted to activate the dynamic identity credential. Once a user activates their dynamic identity credentials, a dynamic risk assessment model is used to score the risk of the basic information and determine the risk level. If the dynamic identity credential does not exceed the predetermined time, the basic information will be risk-scored and the risk level will be determined through a dynamic risk assessment model.

[0044] This embodiment ensures that the identity credentials relied upon for subsequent risk assessments are valid through a pre-verification of validity period. This avoids security risks and data accuracy issues that may arise from risk assessments based on expired or invalid credentials.

[0045] Specifically, in the non-first-time real-name registration process, after the system successfully obtains the basic information submitted by the user (including the vehicle VIN code, vehicle network card ICCID, and dynamic identity certificate), the system does not immediately trigger the risk assessment model. Instead, it first performs a key preparatory step, namely, verifying the validity period of the dynamic identity certificate.

[0046] The system extracts the dynamic identity credential and its associated timestamp from the data packet submitted by the user. Then, the system compares the timestamp in the credential with the current standard time of the server, calculating the time difference. The system presets a valid time window (i.e., a predetermined time), such as 24 hours. By determining whether the time difference exceeds the valid time window, the system accurately verifies whether the dynamic identity credential has expired.

[0047] This verification step creates a clear branching path. If the system determines that the dynamic identity credential has exceeded its predetermined time, the subsequent risk assessment process will not be initiated. Instead, the system will send a clear prompt message to the user's client, guiding the user to reactivate the dynamic identity credential. The activation process typically requires the user to provide some core identity information again (such as ID card number and mobile phone number) and obtain a new SMS verification code to complete the credential update.

[0048] Only after the user successfully activates their credential as prompted will the system trigger the dynamic risk assessment model based on the updated and valid credential status. This model will score the risk of the basic information and related context of the registration request and determine its risk level.

[0049] Conversely, if the system determines during the initial verification that the dynamic identity credential is within its validity period and has not exceeded the predetermined time, the process will proceed seamlessly. The system will immediately invoke the dynamic risk assessment model to directly score and determine the risk level of the current request, thus proceeding to the subsequent differentiated processing stage.

[0050] This pre-verification mechanism ensures that the risk assessment model is activated only if the credentials are valid, which not only guarantees the reliability of the security assessment basis but also optimizes the scheduling of system resources.

[0051] In this embodiment, please refer to Figure 3 The dynamic risk assessment model is used to score the risk of basic information and determine the risk level, including: In offline mode, static risk parameters are collected, including historical real-name registration records, VIN code binding records, SIM card binding records, and SIM card activation status. The static risk parameters are scored and assigned values ​​based on the preset static quantification standards to obtain the static risk score. During the real-name registration request, dynamic risk parameters are monitored in real time. These dynamic risk parameters include registration during off-peak business hours, abnormal IP geographical location, high-frequency requests in a short period of time, and abnormal time consumption of key steps. Based on a preset dynamic quantification standard, each dynamic risk parameter is scored and assigned a value to obtain a dynamic risk score. The analytic hierarchy process (AHP) is used to assign weights to static and dynamic risk scores, calculate a comprehensive risk score, and determine the risk level based on the comprehensive risk score.

[0052] This embodiment employs the Analytic Hierarchy Process (AHP) to weight static and dynamic risk scores, reflecting both the long-term stability of static parameters (e.g., historical real-name registration records account for 50% of the weight) and the immediate sensitivity of dynamic parameters (e.g., abnormal IP addresses account for 30% of the weight). This method uses multi-dimensional indicators for quantification, avoiding the dominance of a single factor in risk assessment and reducing the false positive rate.

[0053] Specifically, when the user's device is offline, the system obtains the user's historical real-name registration records (such as the number of registrations, historical manual review results, abnormal biometric verification, number of real-name cards, etc.), vehicle VIN code binding records (such as frequent binding of VIN codes to vehicle networking cards, which may indicate the risk of vehicle theft), SIM card binding records (such as frequent changes of SIM cards to different devices, which may indicate the risk of card theft), and SIM card status (whether there are any bad records such as arrears or abnormal suspension). Next, the above parameters are scored and assigned values ​​based on the preset static quantitative standards. For example, 2 points are deducted for each additional failure in the historical real-name registration record, and 1 point is added for each additional success. If the binding time of the VIN code binding record exceeds 6 months, 3 points are added. If the SIM card binding record has multiple device bindings, 5 points are deducted. If the SIM card activation status is not activated, 10 points are deducted. Finally, a static risk score is generated. Upon receiving a real-name registration request, the system obtains information such as the IP address, geographical location, request time, and request frequency of the real-name registration request in real time, based on the user's real-name registration request. At the same time, it monitors abnormal behavior in real time, including real-name registration during non-peak business hours, such as a large number of real-name registration operations in the early morning; the location of the real-name registration request IP address does not match the vehicle registration address and the actual area of ​​use; frequent real-name registration requests in a short period of time (such as ≥5 attempts to supplement registration within 30 minutes); and abnormal time consumption of key steps in real-name registration, such as the time for filling in identity information being too short or too long. Next, based on the preset dynamic quantification standard, the dynamic parameters are scored and assigned values. For example, 3 points are deducted for registration during non-peak business hours, 5 points are deducted for abnormal IP geographical location, 2 points are deducted for each high-frequency request in a short period of time, and 4 points are deducted for abnormal time consumption of key steps. The results are then used to generate a dynamic risk score. Then, the Analytic Hierarchy Process (AHP) is used to construct a weighting model for static and dynamic risk scores. The weights of static and dynamic risk scores are determined by constructing a judgment matrix and calculating eigenvectors. For example, the weight of static risk score is 50%, and the weight of dynamic risk score is 50%. Among the static risks, the weights of historical real-name registration records, VIN code binding history, SIM card binding history, and SIM card activation status are 30%, 20%, 20%, and 30%, respectively. Among the dynamic risks, the weights of real-name registration during non-peak business hours, abnormal IP geographical location, frequent real-name registration initiation in a short period of time, and abnormal time consumption of key steps in real-name registration are 20%, 30%, 30%, and 20%, respectively.

[0054] The static risk score and the dynamic risk score are then weighted and summed according to their respective weights to obtain a comprehensive risk score. If the score is less than or equal to 40, it is judged as low risk; if the score is greater than 41 and less than or equal to 70, it is judged as medium risk; and if the score is greater than 70, it is judged as high risk. The corresponding handling strategy is then triggered according to the risk level.

[0055] S107: Implement the corresponding registration process according to different risk levels.

[0056] Specifically, the registration process, which is carried out according to different risk levels, includes: If the risk level is low, the secondary verification and manual review will be skipped, and the basic information will be directly encrypted and transmitted to the real-name registration backend system to complete the real-name registration. If the risk level is low, then a second verification will be performed; If the risk level is high, the process will proceed to manual review.

[0057] By precisely matching risk levels with differentiated processing strategies, efficiency is maximized while ensuring security. Low-risk users enjoy a fast registration experience, medium-risk users undergo a moderate verification process to ensure security, and high-risk users are subject to strict manual review, forming a tiered security protection system.

[0058] More specifically, if the system determines that the current request has a low risk level, it automatically enters the fast registration channel. This system skips all secondary verification and manual review steps, directly transmitting the basic registration information, including the vehicle VIN code, vehicle network card ICCID, and associated encrypted identity data packet, to the backend real-name registration system via a secure link. After receiving the data, the backend system completes the information binding and immediately returns a registration success result, achieving fully automated, second-level completion.

[0059] If the system determines the risk level to be medium, it triggers an enhanced verification process. The system automatically calls the integrated public security population information database's facial comparison service interface, guiding the user to complete a 1:1 facial recognition verification. Verification is only considered successful when the similarity reaches or exceeds a preset threshold of 90%. After successful verification, the system transmits the registration information to the backend to complete real-name registration.

[0060] If the system determines the risk level to be high, the request is automatically routed to the manual review queue. The system pushes all relevant information about the request, including the risk scoring criteria, the user's submitted basic information, and associated historical records, to the workstation of a professional customer service representative. The representative then conducts a substantive review of the registration request and makes a final decision to approve or reject it based on their professional judgment. All paths ultimately converge on the final step of completing real-name registration.

[0061] In some embodiments, it also includes: Real-time collection of multi-dimensional data of current real-name registration requests, including user identity information, device fingerprint information, network address information, vehicle identification code, and request timestamp; User identity information, device fingerprint information, network address information, and vehicle identification code are abstracted as graph nodes, and real-name registration requests are abstracted as edges connecting graph nodes. Each edge is assigned an attribute containing the request timestamp to obtain a graph network. The community detection algorithm is applied to the graph network to identify the node communities connected in the topology and to extract dynamic features of the node communities. The dynamic features include the growth rate of the number of nodes in the node community, the frequency of the same device or network address associated with different identities in the node community, and the consistency of the behavior patterns exhibited by the node community in a short period of time. The dynamic features are matched and risk scored in real time with a pre-set database of group fraud patterns. When the risk score of a node community exceeds a preset threshold, the node community is determined to be a high-risk community. Implement security blocking strategies for all current and subsequent real-name registration requests belonging to high-risk groups.

[0062] This embodiment, by updating the graph network in real time and extracting dynamic features of the community, enables the system to capture the evolutionary patterns of fraudulent behavior over time. Features such as the growth rate of node numbers and the consistency of behavioral patterns allow the system to promptly detect emerging fraud groups, rather than waiting for fraudulent activity to occur before responding. Simultaneously, the community detection algorithm can automatically identify potential risk groups from complex relationships. This unsupervised learning method overcomes the limitation of traditional rule engines that require pre-defined fraud patterns. The system can not only identify known fraud patterns but also discover novel and unknown group fraud characteristics.

[0063] Specifically, when a user initiates a real-name registration request for a vehicle network card, the system collects multi-dimensional data in real time: extracting user identity information such as name and ID number from the registration form submitted by the user; obtaining the device's unique fingerprint information (including device model and chip serial number) through the hardware identification module built into the vehicle network device; collecting network address information such as the IP address and location of the registration request through the network access module; obtaining the vehicle identification number (VIN) from user input or vehicle OCR recognition results; and synchronously generating a system timestamp (accurate to milliseconds) when the registration request is initiated.

[0064] The system abstracts the collected multi-dimensional data into a graph network structure: user identity information (uniquely identified by ID card number), device fingerprint information (uniquely identified by chip serial number), network address information (uniquely identified by IP address), and vehicle identification code (uniquely identified by VIN code) are each used as an independent graph node; a single real-name registration request is abstracted as an undirected edge connecting the above nodes, and the corresponding request timestamp is written into the attribute field of each edge, forming a graph network containing nodes, edges, and edge attributes.

[0065] The system uses the Louvain community discovery algorithm to analyze the graph network. By calculating the connection density between nodes and the weight of edges (based on request timestamp similarity), it identifies tightly connected node clusters in the topology. For example, a cluster containing 5 user identity nodes, 2 device fingerprint nodes, 1 network address node, and 3 vehicle identification code nodes is identified, with all nodes in the cluster interconnected through edges related to real-name registration requests. The system then performs dynamic feature extraction on this node cluster: it counts the number of nodes in the cluster increasing from the initial 3 to 18 within one hour, calculating a growth rate of 15 nodes per hour; it analyzes the 2 device fingerprint nodes, noting that 1 device fingerprint is associated with 4 different user identity nodes, and 1 network address node is associated with 5 different user identity nodes, thus determining the frequency of the same device / network address associating with different identities; and it analyzes the timestamps of all registration requests within the cluster, finding that they are concentrated between 0:00 and 2:00 daily, determining a 90% consistency in their behavioral patterns.

[0066] The system calls a pre-defined feature library of group fraud patterns. This feature library includes features such as a node growth rate of ≥10 nodes / hour, the frequency of the same device / network address being associated with different identities ≥3 times, and behavioral pattern consistency ≥80%, along with corresponding scores (30 points for each matching, out of a maximum of 90 points). The extracted dynamic features are matched against the feature library item by item. If the node cluster matches all three features, it scores 90 points. The pre-defined risk scoring threshold is 80 points. If the cluster's score exceeds the threshold, it is determined to be a high-risk cluster.

[0067] The system marks all nodes within a high-risk cluster (including 5 user identity nodes, 2 device fingerprint nodes, 1 network address node, and 3 vehicle identification code nodes). For real-name registration requests currently being initiated from these nodes, the system directly returns a registration failure message and informs the user of the risk. For real-name registration requests initiated by any node within the cluster within the next 24 hours, the system automatically intercepts and rejects them, and simultaneously synchronizes the interception record to the risk control log for subsequent source tracing and analysis.

[0068] In some embodiments, the method for constructing a group fraud pattern feature database includes: collecting confirmed group fraud case data from historical periods, which contains complete fraud operation chains involving multiple related real-name registration request records and their topological relationships in a graph network. The data preprocessing stage cleans and standardizes the raw data to ensure the accuracy of subsequent analysis.

[0069] Next, based on the cleaned data, the system performs multi-dimensional feature extraction. The extracted features cover structural features, including the node size, edge connection density, and node type diversity of the fraud community; temporal features, including the distribution pattern of operation time, request initiation frequency pattern, and behavioral sequence correlation; and relational features, including the complexity of device and identity association, the degree of network address clustering, and vehicle identification code usage patterns.

[0070] The system then employs an unsupervised learning algorithm to perform cluster analysis on the extracted features. By combining density clustering and hierarchical clustering, it identifies fraud groups with similar patterns. Each cluster center forms an initial fraud pattern template, which contains the key feature vectors of that pattern and their numerical ranges.

[0071] A risk assessment system is then established for each fraud pattern template. Based on dimensions such as the historical frequency of occurrence of the pattern, the actual extent of loss caused, and the difficulty of detection, the system assigns basic risk weights and matching thresholds to the templates through a combination of expert scoring and machine learning. Templates with higher weights represent more threatening fraud patterns.

[0072] After the feature library is established, it enters a continuous optimization phase. The system is equipped with a model update mechanism that regularly collects newly occurring fraud cases and updates existing template features and weights through incremental learning algorithms. At the same time, concept drift detection technology is introduced. When a significant change in fraud patterns is detected, the template reconstruction process is automatically triggered to ensure that the feature library can adapt to constantly changing fraud methods.

[0073] The completed feature library of group fraud patterns is integrated with a real-time analysis engine, providing a benchmark comparison standard for graph network analysis. The feature library, through versioned management, supports the tracking and research of fraud patterns from different periods, forming a complete knowledge accumulation system.

[0074] In some embodiments, the method further includes: collecting multi-dimensional data of real-name registration requests, including user identity information, device fingerprint information, network address information, vehicle identification code and request timestamp, and constructing a dynamic association graph containing four types of entity nodes (identity, device, network and vehicle) in real time through a distributed stream processing framework. Next, the dynamic association graph is input into the graph attention network (GAT). Weights are assigned to different types of edges through the attention mechanism. The initial weight values ​​of directly associated edges (such as device-identity, identity-vehicle) are higher than those of indirectly associated edges (such as identity-network-device). The GAT is then trained under supervision based on historical fraud case data, so that the network can automatically learn fraud pattern features in multi-level association paths (such as device A→identity B→network C→device D). Then, the node embedding vectors output by GAT are extracted, the cross-level association strength of each edge is calculated, and the complex association topology is analyzed by weighted adjacency matrix to identify suspicious node clusters containing at least three layers of indirect association (such as device A indirectly associated with device D through identity B and network C). The cross-level association features of the suspicious node clusters are then matched with a preset fraud pattern feature library. The feature matching rules are dynamically adjusted through the hidden layer parameters of the graph neural network to generate a risk score. The score includes the cross-level association depth weight (5 points are added for each additional layer of indirect association) and the edge weight decay coefficient (the indirect edge weight decays exponentially by 0.8). Subsequently, when the risk score of the community exceeds the preset threshold, a security interception strategy is triggered, which routes the current and subsequent real-name registration requests of all nodes in the community to the manual review queue, and simultaneously records the interception reason, associated path and risk score to the blockchain evidence storage chain. After the manual review result is confirmed, the fraud pattern feature library is updated and the GAT model is retrained.

[0075] This embodiment assigns high weights to directly related edges such as devices and identities through an attention mechanism, while the weights of indirectly related edges such as identities, networks, and devices decay hierarchically (e.g., the weight of an indirect related edge is multiplied by 0.8 for each additional layer), enabling the model to accurately distinguish the importance of direct and indirect relationships. This effectively filters out noisy associations.

[0076] Specifically, multi-dimensional data from real-name registration requests is collected, including user identity information (name, ID number, mobile phone number), device fingerprint information (IMEI code, operating system version), network address information (IP geolocation, carrier affiliation), vehicle identification number (VIN code), and request timestamp. A dynamic association graph containing four types of entity nodes (identity, device, network, and vehicle) is constructed in real time using a distributed stream processing framework (such as Apache Flink). Each type of entity node establishes direct edges through shared fields (such as associating the same ID number with the identity and device nodes), while indirect edges are generated through multiple paths (such as identity-network-device). Next, the dynamic association graph is input into the graph attention network (GAT). Initial weights are assigned to different types of edges through the attention mechanism. The initial weight value of direct association edges (such as device-identity, identity-vehicle) is set to 0.9, and the initial weight value of indirect association edges (such as identity-network-device) is set to 0.5. The GAT is trained under supervision based on historical fraud case data, so that the network can automatically learn the fraud pattern features in multi-level association paths (such as device A→identity B→network C→device D). During the training process, the attention weight parameters are optimized through backpropagation, so that the weights of fraud association paths gradually converge to higher values. Then, the node embedding vectors output by GAT are extracted, the cross-level association strength of each edge is calculated, and the complex association topology is analyzed by weighted adjacency matrix to identify suspicious node clusters containing at least three layers of indirect association (such as device A indirectly associated with device D through identity B and network C). The formula for calculating the cross-level association strength is: association strength = Σ (edge ​​weight × 2^(-number of levels)), where the number of levels refers to the number of edges in the path. The cross-level association features of suspicious node clusters are then matched with a pre-defined fraud pattern feature library. The feature matching rules are dynamically adjusted using the hidden layer parameters of the graph neural network to generate a risk score. The score includes a weight for the depth of cross-level associations (adding 5 points for each additional layer of indirect association) and an edge weight decay coefficient (indirect edge weights decay exponentially by 0.8). For example, the score for the path Device A → Identity B → Network C → Device D is calculated as: 5 × 3 (three layers of indirect association) + (0.9 + 0.5 × 0.8 + 0.5 × 0.8). 2 ) = 15 + 1.62 = 16.62; When the risk score of a cluster exceeds a preset threshold (e.g., 15 points), a security interception strategy is triggered. All current and subsequent real-name registration requests of all nodes in the cluster are routed to the manual review queue. The interception reason (e.g., three-layer indirect association + 16.62 points), association path (device A → identity B → network C → device D) and risk score are recorded in the blockchain evidence storage chain. After the manual review result is confirmed, if it is marked as fraud, the cross-level association features of the cluster are updated to the fraud pattern feature library, and the GAT model is incrementally trained based on the new features to optimize the attention weight distribution and scoring rules.

[0077] In some embodiments, during the real-name registration process, the operational status of core dependent components is monitored in real time. These core dependent components include a real-time risk assessment model, a blockchain network, and a facial recognition interface. When any component experiences a communication timeout, response error, or service unavailability failure, a tiered degradation strategy is automatically executed according to preset rules. The tiered degradation strategy includes: when the real-time risk assessment model is unavailable, suspending the collection and analysis of dynamic risk parameters and switching to a simplified risk assessment mode based on static risk parameters. This simplified mode calculates a static risk score based on the user's historical real-name records, VIN code, and SIM card binding status, and maps this score to a risk level. When the blockchain network is congested or experiences excessive latency, the real-time risk assessment model is suspended. For on-chain data storage, the data to be stored is temporarily written to a high-speed cache queue and marked as pending synchronization. Once the blockchain network recovers, the data is automatically read from the cache queue and the delayed on-chain storage is completed. When the face recognition interface is unavailable or the response times out, requests with a medium risk level are automatically upgraded to a high-risk processing path and routed to a manual review queue, where customer service personnel verify the identity based on other credentials. All operations triggered by the degradation strategy, the reasons for degradation, and the original data are recorded in the system audit log. During the execution of the degradation strategy, the health status of the faulty components is continuously polled. Once the component service is detected to have recovered, the system automatically switches back to the standard registration process and simultaneously executes the operations that were temporarily suspended due to the strategy adjustment during the degradation period.

[0078] Specifically, during the operation of the vehicle network card real-name registration system, the system continuously monitors the operational status of its core dependent components in real time. These core components include a real-time risk assessment model, blockchain network services, and a facial recognition interface. The system uses methods such as heartbeat detection, response time analysis, and error code monitoring to promptly detect fault states such as communication timeouts, response errors, or service unavailability in these components.

[0079] When the system detects that the real-time risk assessment model service is unavailable, it automatically triggers a tiered degradation strategy. The system suspends the collection and analysis of dynamic risk parameters and switches to a simplified risk assessment mode based on static risk parameters. This simplified mode uses a predefined rule engine to comprehensively analyze the user's historical real-name registration records, VIN code binding records, and SIM card activation status to calculate a static risk score. This score is then mapped to the corresponding risk level to continue supporting subsequent business judgments.

[0080] When the system detects congestion or excessive latency in the blockchain network, it automatically suspends real-time data upload operations. The system temporarily writes the real-name registration data to be stored into a high-speed cache queue and marks each data entry as pending synchronization. Once the blockchain network returns to normal, the system automatically reads the data from the cache queue in chronological order and performs delayed upload operations to ensure data integrity and traceability.

[0081] When the facial recognition interface service is unavailable or times out, the system automatically adjusts its business logic. For all registration requests classified as medium risk, the system automatically escalates their processing path to high risk, routing them directly to the manual review queue. Human customer service personnel then complete the identity verification and approval process based on other identity verification materials submitted by the user.

[0082] The system records all operations triggered by the degradation policy, the reasons for the degradation, and the corresponding original business data in the system audit log, forming a complete operation trajectory. During the execution of the degradation policy, the system continuously polls the health status of the faulty components and checks the service recovery status through a periodic retry mechanism. Once the system detects that the faulty component service has returned to normal, it immediately and automatically switches back to the standard registration process and simultaneously executes all operations that were suspended due to the policy adjustment during the degradation period, ensuring the eventual consistency of business data.

[0083] This embodiment simplifies the secondary real-name registration process to three basic information submissions through a dynamic identity credential reuse mechanism, reducing the amount of information entered by 70% compared to traditional methods. The automatic approval rate for low-risk users reaches over 80%, and the real-name registration time is reduced from an average of 10 minutes to within 30 seconds. Simultaneously, through a dynamic risk assessment model and differentiated layered security processing, abnormal real-name registration requests are intercepted in real time, achieving a risk identification accuracy rate of over 98%. The dynamic identity credential uses MD5+dynamic salt encryption, combined with timestamp validity control, effectively preventing the malicious interception and reuse of identity information.

[0084] Please see Figure 4 and Figure 5 This embodiment provides a vehicle network card real-name registration device 200, including: The judgment unit 201 is used to obtain the user's real-name registration request and determine whether the real-name registration is the first real-name registration based on the real-name registration request; The generation unit 202 is used to collect the user's identity information and SMS verification code and generate a dynamic identity credential based on the identity information and SMS verification code if the real-name registration is the first real-name registration. Encryption unit 203 is used to encrypt and store the identity information using the dynamic identity credential as the key; Registration unit 204 is used to complete real-name registration by calling the real-name registration interface based on the encrypted identity information; The acquisition unit 205 is used to acquire the user's basic information if the real-name registration is not the first real-name registration; Scoring unit 206 is used to score the basic information and determine the risk level through a dynamic risk assessment model; Execution unit 207 is used to execute the corresponding registration process according to different risk levels.

[0085] Furthermore, the generation unit 202 includes: The data collection subunit is used to collect user name, ID number, on-site photo, liveness video, mobile phone number, timestamp, and SMS verification code; The splicing subunit is used to use the SMS verification code as a dynamic salt and employ a hash algorithm to splice the user's name, ID number, mobile phone number, timestamp, and SMS verification code to obtain a dynamic identity credential.

[0086] Furthermore, the basic information includes: vehicle VIN code, vehicle network card ICCID, and dynamic identity certificate.

[0087] Furthermore, the scoring unit 206 includes: The time verification subunit is used to verify whether the dynamic identity credential has exceeded a predetermined time. The activation subunit is used to remind the user to activate the dynamic identity credential if the dynamic identity credential exceeds a predetermined time. The first-level confirmation subunit is used to score the risk of the basic information and determine the risk level through the dynamic risk assessment model after the user activates the dynamic identity credential. The second-level confirmation subunit is used to score the risk of the basic information and determine the risk level through the dynamic risk assessment model if the dynamic identity credential has not exceeded the predetermined time.

[0088] Furthermore, the scoring unit 206 includes: The static parameter acquisition subunit is used to acquire static risk parameters in an offline state. The static risk parameters include historical real-name registration records, VIN code binding records, SIM card binding records, and SIM card activation status. Static scoring units are used to score and assign values ​​to each static risk parameter based on a preset static quantification standard to obtain a static risk score. The dynamic parameter acquisition subunit is used to monitor dynamic risk parameters in real time when a real-name registration request is made. The dynamic risk parameters include registration during non-peak business hours, abnormal IP geographical location, high-frequency requests in a short period of time, and abnormal time consumption of key steps. Dynamic molecular unit scoring is used to score and assign values ​​to each dynamic risk parameter based on a preset dynamic quantification standard to obtain a dynamic risk score. The weight allocation subunit is used to assign weights to the static risk score and the dynamic risk score using the analytic hierarchy process (AHP), calculate the comprehensive risk score, and determine the risk level based on the comprehensive risk score.

[0089] Furthermore, the execution unit 207 includes: The skip sub-unit is used to skip the secondary verification and manual review if the risk level is low, and directly transmit the basic information to the real-name registration backend system in encrypted form to complete the real-name registration. The secondary verification subunit is used to perform secondary verification if the risk level is low risk. The manual review subunit is used to initiate the manual review process if the risk level is high.

[0090] Furthermore, it also includes: A multi-dimensional data acquisition unit is used to collect multi-dimensional data of the current real-name registration request in real time. The multi-dimensional data includes user identity information, device fingerprint information, network address information, vehicle identification code, and request timestamp. The graph network construction unit is used to abstract the user identity information, device fingerprint information, network address information, and vehicle identification code into graph nodes, and to abstract the real-name registration request into edges connecting the graph nodes. Each edge is assigned an attribute containing the request timestamp to obtain the graph network. The feature extraction unit is used to perform a community detection algorithm on the graph network, identify the node clusters connected on the topology, and perform dynamic feature extraction on the node clusters. The dynamic features include the growth rate of the number of nodes in the node cluster, the frequency of the same device or network address associated with different identities in the node cluster, and the consistency of the behavior patterns exhibited by the node cluster in a short period of time. The matching unit is used to perform real-time matching and risk scoring of the dynamic features with a preset group fraud pattern feature library; The judgment unit is used to determine that a node community is a high-risk community when the risk score of a certain node community exceeds a preset threshold. The interception unit is used to execute security interception strategies on all current and subsequent real-name registration requests belonging to the high-risk group.

[0091] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described apparatus and unit can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0092] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed, can implement the methods provided in the above embodiments. The storage medium may include various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0093] The present invention also provides a computer device, which may include a memory and a processor. The memory stores a computer program, and when the processor calls the computer program in the memory, it can implement the methods provided in the above embodiments. Of course, the computer device may also include various network interfaces, power supplies, and other components.

[0094] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section. It should be noted that those skilled in the art can make various improvements and modifications to this invention without departing from its principles, and these improvements and modifications also fall within the protection scope of the claims of this invention.

[0095] It should also be noted that, in this specification, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusivity.

[0096] The term "comprises" implies that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprises a..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

Claims

1. A method for real-name registration of a vehicle network card, characterized in that, include: Obtain the user's real-name registration request, and determine whether the real-name registration is the first real-name registration based on the real-name registration request; If the real-name registration is the first real-name registration, the user's identity information and SMS verification code are collected, and a dynamic identity credential is generated based on the identity information and SMS verification code. The identity information is encrypted and stored using the dynamic identity credential as the key; Real-name registration is completed by calling the real-name registration interface based on the encrypted identity information; If the real-name registration is not the first time, then the user's basic information will be obtained; The basic information is scored and its risk level is determined using a dynamic risk assessment model. The corresponding registration process will be carried out according to different risk levels.

2. The method for real-name registration of vehicle network cards according to claim 1, characterized in that, The process of collecting user identity information and SMS verification codes, and generating dynamic identity credentials based on the identity information and SMS verification codes, includes: Collect user name, ID number, on-site photo, live video, mobile phone number, timestamp, and SMS verification code; Using the SMS verification code as a dynamic salt, a hash algorithm is used to concatenate the user's name, ID number, mobile phone number, timestamp, and SMS verification code to obtain a dynamic identity credential.

3. The method for real-name registration of vehicle network cards according to claim 1, characterized in that, Basic information includes: vehicle VIN code, vehicle network card ICCID, and dynamic identity certificate.

4. The method for real-name registration of vehicle network cards according to claim 3, characterized in that, Before performing risk scoring and determining the risk level of the basic information using a dynamic risk assessment model, the following steps are included: Verify whether the dynamic identity credential has exceeded the predetermined time; If the dynamic identity credential exceeds the predetermined time, the user will be prompted to activate the dynamic identity credential. When a user activates the dynamic identity credential, the basic information is scored and the risk level is determined by the dynamic risk assessment model. If the dynamic identity credential does not exceed the predetermined time, the basic information is risk-scored and the risk level is determined by the dynamic risk assessment model.

5. The method for real-name registration of vehicle network cards according to claim 1, characterized in that, The step of scoring the basic information and determining the risk level using a dynamic risk assessment model includes: In offline mode, static risk parameters are collected, including historical real-name registration records, VIN code binding records, SIM card binding records, and SIM card activation status; The static risk parameters are scored and assigned values ​​based on the preset static quantification standards to obtain the static risk score. During the real-name registration request, dynamic risk parameters are monitored in real time. These dynamic risk parameters include registration during non-peak business hours, abnormal IP geographical location, high-frequency requests in a short period of time, and abnormal time consumption of key steps. Based on a preset dynamic quantification standard, each dynamic risk parameter is scored and assigned a value to obtain a dynamic risk score. The static risk score and dynamic risk score are weighted using the analytic hierarchy process (AHP) to calculate a comprehensive risk score, and the risk level is determined based on the comprehensive risk score.

6. The method for real-name registration of vehicle network cards according to claim 1, characterized in that, The process of performing corresponding registration procedures based on different risk levels includes: If the risk level is low, then skip the secondary verification and manual review, and directly encrypt and transmit the basic information to the real-name registration backend system to complete the real-name registration. If the risk level is low, then a second verification is performed; If the risk level is high, then the manual review process will begin.

7. The method for real-name registration of vehicle network cards according to claim 1, characterized in that, Also includes: The system collects multi-dimensional data of the current real-name registration request in real time. The multi-dimensional data includes user identity information, device fingerprint information, network address information, vehicle identification code, and request timestamp. The user identity information, device fingerprint information, network address information, and vehicle identification code are abstracted as graph nodes, and the real-name registration request is abstracted as an edge connecting the graph nodes. Each edge is assigned an attribute containing the request timestamp to obtain a graph network. The community detection algorithm is applied to the graph network to identify the node clusters connected in the topology, and dynamic features are extracted from the node clusters. The dynamic features include the growth rate of the number of nodes in the node cluster, the frequency of the same device or network address associated with different identities in the node cluster, and the consistency of the behavior patterns exhibited by the node cluster in a short period of time. The dynamic features are matched and risk scored in real time with a preset group fraud pattern feature library; When the risk score of a node community exceeds a preset threshold, the node community is determined to be a high-risk community. A security blocking strategy is implemented for all current and subsequent real-name registration requests belonging to the aforementioned high-risk groups.

8. A vehicle network card real-name registration device, characterized in that, include: The judgment unit is used to obtain the user's real-name registration request and determine whether the real-name registration is the first real-name registration based on the real-name registration request; The generation unit is used to collect the user's identity information and SMS verification code and generate a dynamic identity credential based on the identity information and SMS verification code if the real-name registration is the first real-name registration. An encryption unit is used to encrypt and store the identity information using the dynamic identity credential as a key. The registration unit is used to complete real-name registration by calling the real-name registration interface based on encrypted identity information. The acquisition unit is used to acquire the user's basic information if the real-name registration is not the first real-name registration; The scoring unit is used to score the basic information based on a dynamic risk assessment model and determine the risk level. The execution unit is used to perform the corresponding registration process according to different risk levels.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the vehicle network card real-name registration method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, causes the processor to perform the vehicle network card real-name registration method as described in any one of claims 1 to 7.

Images