Multi-modal education data security processing method and system based on national cryptographic algorithm and homomorphic encryption

Abstract

The invention discloses a multi-modal education data security processing method and system based on a national cryptographic algorithm and homomorphic encryption, and belongs to the technical field of data security. The method comprises the following steps: acquiring multi-modal education data, classifying the multi-modal education data, preprocessing various types of data, generating an encryption key by using a national cryptographic algorithm, encrypting the preprocessed data by combining a homomorphic encryption technology, generating a corresponding ciphertext, executing homomorphic calculation operation on the ciphertext from a plurality of data sources in an encryption state, and generating a homomorphic calculation result. Data analysis of privacy protection is realized, a user is authorized to decrypt a calculation result ciphertext by using a corresponding private key, the integrity of a decryption result and the authenticity of a data source are verified by using a national secret algorithm, a data access authority is managed by using a national secret identity-based encryption algorithm, encrypted data are safely stored, and access control based on identities is implemented. According to the method, national secret and homomorphic encryption are fused, and safe and efficient processing of education data is realized.

Description

Technical Field

[0001] This invention relates to the field of data security technology, specifically to a method and system for secure processing of multimodal educational data based on national cryptographic algorithms and homomorphic encryption. Background Technology

[0002] With the deepening of educational informatization, applications such as smart education, online learning platforms, and educational big data analytics are rapidly developing, fundamentally transforming the ways in which educational data is collected, processed, and utilized. Current educational data exhibits significant multimodal characteristics: it includes structured data such as student registration information and grade records, unstructured data such as teaching videos and audio courseware, and time-series streaming data such as classroom environment monitoring data from IoT devices. This multimodal data plays an increasingly important role in supporting personalized teaching, educational quality assessment, and educational management decision-making.

[0003] However, educational data often contains a large amount of sensitive information, such as student personal privacy, academic performance, and teaching evaluation results, posing a serious challenge to its security. Current technologies primarily employ the following solutions for the secure processing of educational data: First, there are data protection schemes based on traditional encryption algorithms (such as AES and RSA). While these schemes can ensure the confidentiality of data during storage and transmission, the encrypted data cannot be directly used for computation and analysis. When data statistics and analysis are required, the data must be decrypted first. This not only increases the risk of data leakage but also leads to additional computational overhead and time delays, making it difficult to meet the real-time requirements of data analysis in educational settings.

[0004] Second, privacy-preserving computation schemes based on homomorphic encryption. Homomorphic encryption allows computation directly in ciphertext, resolving the conflict between privacy protection and data utilization to some extent. However, existing homomorphic encryption schemes suffer from significant performance bottlenecks: the ciphertext inflation rate is typically more than twice that of the original data, significantly increasing storage and transmission overhead; decryption latency often exceeds 200 milliseconds, making it difficult to meet the interactive analysis needs of educational applications. These performance issues are particularly pronounced when processing large-scale unstructured data such as video.

[0005] Third, there are security solutions for single data types. Existing technologies mostly focus on the security protection of a specific type of data, lacking a unified framework for processing multimodal educational data. For example, some solutions are specifically designed for structured databases, while others focus on multimedia data protection. This fragmented approach leads to complex educational data security management systems, high operation and maintenance costs, and difficulty in achieving cross-modal data correlation analysis.

[0006] Furthermore, in scenarios involving the sharing and collaborative analysis of educational data, such as federal statistics among multiple educational institutions and cross-regional educational quality assessments, existing solutions have significant shortcomings. Traditional methods require data anonymization before sharing, but the value of the anonymized data is greatly reduced; while data encryption-based methods struggle to support secure multi-party computation.

[0007] To address the aforementioned issues, there is an urgent need for a multimodal educational data security processing method and system based on national cryptographic algorithms and homomorphic encryption to solve the problems existing in traditional methods. Summary of the Invention

[0008] The purpose of this invention is to provide a method and system for secure processing of multimodal educational data based on national cryptographic algorithms and homomorphic encryption. By integrating national cryptographic algorithms and homomorphic encryption technology, it provides high-strength security protection for multimodal educational data while supporting efficient computation and analysis in ciphertext state. This resolves the contradiction between privacy protection and data utilization, and significantly reduces ciphertext expansion and computational latency, providing a feasible technical path for secure sharing and collaborative analysis of educational data.

[0009] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A method for securely processing multimodal educational data based on national cryptographic algorithms and homomorphic encryption includes: Step 1: Acquire multimodal educational data, classify it according to data type, and preprocess each type of data to convert it into a unified format suitable for encrypted computing. The multimodal educational data includes structured data, unstructured data, and time-series streaming data. Step 2: Use the national cryptographic algorithm to generate an encryption key, and combine it with homomorphic encryption technology to encrypt the preprocessed data to generate the corresponding ciphertext; Step 3: In an encrypted state, perform homomorphic computation operations on ciphertext from multiple data sources to achieve privacy-preserving data analysis; Step 4: Authorize users to decrypt the ciphertext of the calculation result using the corresponding private key, and use the national cryptographic algorithm to verify the integrity of the decryption result and the authenticity of the data source; Step 5: Use the national cryptographic identity base encryption algorithm to manage data access permissions, securely store encrypted data, and implement identity-based access control.

[0010] Furthermore, in step 1, the structured data is preprocessed, specifically as follows: Extract the key fields from the structured data and serialize them into binary format.

[0011] Furthermore, in step 1, the unstructured data is preprocessed, specifically as follows: Unstructured data is processed in segments and features are extracted. The feature extraction is implemented using a convolutional neural network.

[0012] Furthermore, in step 1, the time-series stream data is preprocessed, specifically as follows: The time-series stream data is subjected to window segmentation and normalization processing, wherein the window segmentation is based on time window sampling.

[0013] Furthermore, in step 2, the encryption key is generated using the national cryptographic algorithm, specifically as follows: Asymmetric key pair public key generated based on SM2 algorithm PK sm2 and private key SK sm2 Used for digital signatures and key exchange; Calculate the hash value of the data based on the SM3 algorithm: H = SM 3 ( D (), where D represents the original data and H represents the hash value, used for integrity verification; Generate a symmetric key based on the SM4 algorithm K sm4 It is used to encrypt unstructured data and time-series streaming data; Generate the identity base key using the SM9 algorithm: Public Key PK sm9 and private key SK sm9 Used for user-identity-based secure access control; Generate homomorphic encryption key pair: public key PK he and private key SK he Homomorphic encryption schemes employ either the Paillier algorithm or the BFV scheme.

[0014] Furthermore, in step 2, the preprocessed data is encrypted using homomorphic encryption technology to generate the corresponding ciphertext, specifically as follows: For structured data, use public keys. PK he Encrypt; For unstructured data, first use a symmetric key. K sm4 Encrypt it, then use the public key. PK he For symmetric keys K sm4 Encrypt; For time-series streaming data, first use the public key. PK he Perform streaming encryption, encrypting each normalized data point individually.

[0015] Furthermore, in step 3, the homomorphic computation operations include: homomorphic addition operations, homomorphic multiplication operations, and federated statistical process operations.

[0016] Furthermore, in step 4, the authorized user decrypts the ciphertext of the calculation result using the corresponding private key, and verifies the integrity of the decryption result and the authenticity of the data source using the national cryptographic algorithm, specifically as follows: Authorized users can use the corresponding private key to decrypt the ciphertext; The hash value of the decrypted data is calculated based on the SM3 algorithm and compared with the original hash value. If they are the same, the data is considered complete. Based on public key PK sm2 Verify the authenticity of the decrypted data.

[0017] Furthermore, in step 5, the national cryptographic identity base encryption algorithm is used to manage data access permissions, specifically as follows: Data access permissions are managed based on the SM9 algorithm. When a user requests data, the user's identity is verified using the SM9 private key. SK sm9 Generate a decryption token. If the token is valid, allow the user to access the data; otherwise, do not allow access.

[0018] This invention also provides a multimodal educational data security processing system based on national cryptographic algorithms and homomorphic encryption, used to implement the above-mentioned method, including: The preprocessing module is used to acquire multimodal educational data, classify it according to data type, and preprocess structured data, unstructured data, and time-series streaming data respectively to transform them into a unified format suitable for encrypted computing; The encryption module, connected to the preprocessing module, is used to generate an encryption key using the national cryptographic algorithm and to encrypt the preprocessed data using homomorphic encryption technology to generate the corresponding ciphertext. The computing module, which communicates with the encryption module, is used to perform homomorphic computation operations on ciphertext from multiple data sources in an encrypted state, thereby achieving privacy-preserving data analysis. The decryption and verification module communicates with the calculation module and is used by authorized users to decrypt the ciphertext of the calculation result using the corresponding private key, and to verify the integrity of the decryption result and the authenticity of the data source using the national cryptographic algorithm. The secure storage and access control module is connected to the encryption module and the decryption and verification module, respectively. It is used to manage data access permissions using the national cryptographic identity base encryption algorithm, securely store encrypted data, and implement identity-based access control.

[0019] In summary, the present invention has at least one of the following beneficial technical effects: 1. This invention achieves an excellent balance between security and computational efficiency. While ensuring high-strength security protection that complies with national cryptographic standards throughout the entire data lifecycle, it supports direct computation of encrypted data, overcomes the security risks of traditional encryption methods that require decryption computation, and significantly improves the problem of low computational efficiency of existing homomorphic encryption technology.

[0020] 2. This invention effectively breaks through the performance bottleneck of homomorphic encryption. By integrating data preprocessing compression, efficient encoding and hardware acceleration, it successfully controls the ciphertext expansion rate and decryption delay, meeting the stringent real-time requirements of educational data analysis scenarios.

[0021] 3. This invention provides comprehensive multimodal data processing capabilities. By establishing a unified and flexible security processing framework through differentiated preprocessing and encryption strategies for structured, unstructured, and time-series stream data, it can fully address the complex processing needs of diverse and heterogeneous data in educational informatization.

[0022] 4. This invention constructs an end-to-end security protection system from the data source to the access terminal. It comprehensively utilizes multiple national cryptographic algorithms to ensure the confidentiality, integrity, and authenticity of data throughout the entire data processing chain. Furthermore, it implements refined access control based on identity-based encryption, fully meeting the compliance requirements of current data security regulations.

[0023] 5. This invention natively supports privacy-preserving computing collaboration scenarios such as federated computing, enabling multiple educational institutions to conduct secure joint statistical analysis without sharing original plaintext data. This fundamentally resolves the contradiction between data sharing and privacy protection, providing a reliable technical foundation for cross-domain value mining of educational data. Attached Figure Description

[0024] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a system structure diagram of the present invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0026] To facilitate understanding, this invention first explains some of the English abbreviations: SM2: Chinese National Standard Asymmetric Encryption Algorithm; SM3: Chinese National Standard Hash Algorithm; SM4: Chinese National Standard Symmetric Encryption Algorithm; SM9: Chinese National Standard Identity-Based Encryption Algorithm; IoT: Internet of Things; Homomorphic encryption.

[0027] like Figure 1 As shown, this invention provides a method for secure processing of multimodal educational data based on national cryptographic algorithms and homomorphic encryption. This method is implemented using a computing device and specifically includes: Step 1: Acquire multimodal educational data, classify it according to data type, and preprocess each type of data to convert it into a unified format suitable for encrypted computing. The multimodal educational data includes structured data, unstructured data, and time-series streaming data. Step 2: Use the national cryptographic algorithm to generate an encryption key, and combine it with homomorphic encryption technology to encrypt the preprocessed data to generate the corresponding ciphertext; Step 3: In an encrypted state, perform homomorphic computation operations on ciphertext from multiple data sources to achieve privacy-preserving data analysis; Step 4: Authorize users to decrypt the ciphertext of the calculation result using the corresponding private key, and use the national cryptographic algorithm to verify the integrity of the decryption result and the authenticity of the data source; Step 5: Use the national cryptographic identity base encryption algorithm to manage data access permissions, securely store encrypted data, and implement identity-based access control.

[0028] In step 1, multimodal educational data is acquired, categorized according to data type, and preprocessed to transform each type of data into a unified format suitable for encrypted computation. The multimodal educational data includes structured data, unstructured data, and time-series streaming data, specifically: The computing device receives multimodal educational data and classifies and preprocesses it according to data type. Multimodal data includes structured data (such as student registration information and grade records), unstructured data (such as instructional videos and audio), and time-series streaming data (such as classroom environment data collected by IoT sensors). Preprocessing aims to convert the data into a format suitable for encryption and computation, specifically including: 1. Preprocess the structured data: Extract key fields from the structured data (such as student ID and course grades) and serialize them into binary format. The serialization formula is as follows: D struct =Serialize( S ),in, S Representing structured data, D struct This represents the serialized binary data.

[0029] 2. Preprocess unstructured data: The computing device performs block processing (dividing the video into fixed-size blocks) and feature extraction (using a convolutional neural network to generate feature vectors). The feature extraction formula is as follows: F =CNN( V ),in, V This represents unstructured data (such as video frames). F denoted as feature vector, and CNN stands for Convolutional Neural Network.

[0030] 3. Preprocess the time-series stream data, specifically as follows: The computing device performs window segmentation (sampling by time window) and normalization (scaling the data to the range [0, 1]). The normalization formula is as follows: ,in, X t Represents time series data points, N t This represents the normalized data.

[0031] In step 2, the encryption key is generated using the national cryptographic algorithm, specifically as follows: Asymmetric key pair public key generated based on SM2 algorithm PK sm2 and private key SK sm2 Used for digital signatures and key exchange; Calculate the hash value of the data based on the SM3 algorithm: H = SM 3 ( D (), where D represents the original data and H represents the hash value, used for integrity verification; Generate a symmetric key based on the SM4 algorithmK sm4 It is used to encrypt unstructured data and time-series streaming data; Generate the identity base key using the SM9 algorithm: Public Key PK sm9 and private key SK sm9 Used for user-identity-based secure access control; Generate homomorphic encryption key pair: public key PK he and private key SK he Homomorphic encryption schemes employ either the Paillier algorithm or the BFV scheme.

[0032] In step 2, homomorphic encryption is used to encrypt the preprocessed data, generating the corresponding ciphertext, specifically as follows: For structured data, homomorphic public keys can be used directly. PK he Encrypted serialized data D struct The encryption formula is: C struct = Enc he ( PK he , D struct ),in, Enc he Represents a homomorphic encryption function. C struct This represents the encrypted structured ciphertext; For unstructured data, first use the SM4 symmetric key. K sm4 Encrypt raw data or feature vectors F Then use homomorphic public key PK he encryption K sm4 The encryption formula is: C data = Enc sm4 ( K sm4 , F ), C key = Enc he ( PK he , K sm4 ),in, Enc sm4 This refers to the SM4 encryption function.C data This represents the encrypted unstructured data ciphertext. C key This represents the SM4 key used for encryption; For time-series streaming data, use homomorphic public keys. PK he Perform streaming encryption on each normalized data point. N t Separate encryption, the encryption formula is: C stream = Enc he ( PK he , N t ),in, C stream This represents the encrypted time-series ciphertext.

[0033] In step 3, the homomorphic computation operations include: homomorphic addition, homomorphic multiplication, and federated statistical process operations, specifically: Computing devices perform computations in encrypted form, such as federal statistics on grades in educational data. Encrypted computations do not require data decryption, thus protecting privacy. The specific computation process is as follows: 1. Homomorphic addition operation: For multiple encrypted score ciphertexts C 1, C 2, ..., C n (From different educational institutions), the computing device uses the homomorphic addition operator ⊕ to calculate the sum of the ciphertext. C sum The formula is: ; where ⊕ corresponds to the addition operation in homomorphic encryption, such as in the Paillier algorithm, C 1⊕ C 2= C 1× C 2mod n 2 , where n is the public key parameter; 2. Homomorphic multiplication operation: Limited to some homomorphic encryption schemes, it can be used to calculate averages, for example, when calculating the average, the sum is calculated first. C sum Then, use homomorphic scalar multiplication to multiply by the constant 1 / n, as shown in the formula: ;in, This represents a homomorphic scalar multiplication operation; 3. Federal statistical process: Each educational institution uploads encrypted data to a central server (computing device); the server performs aggregation calculations on the encrypted data to generate the resulting encrypted data.C result The encrypted result is returned to the authorized user (such as the education administration department).

[0034] In step 4, the authorized user decrypts the ciphertext of the calculation result using the corresponding private key, and verifies the integrity of the decryption result and the authenticity of the data source using the national cryptographic algorithm. Specifically: 1. Authorized users decrypt the ciphertext using the corresponding private key. The decryption process involves using the homomorphic private key. SK he Decryption result ciphertext C result The decryption formula is: R = Dec he ( SK he , C result ),in, Dec he This represents a homomorphic decryption function. R This indicates a plain text result (such as a statistical score).

[0035] 2. Integrity verification: Recalculate the hash value using the SM3 algorithm. H =SM3 ( R ), and the original hash value H Comparison, if If so, the data is complete.

[0036] 3. Digital signature verification: using the SM2 public key PK sm2 Verify the authenticity of the data source.

[0037] To control decryption latency (<100ms), computing devices employ algorithmic optimizations (such as pre-computation and parallel processing) and hardware acceleration (such as using GPUs). To control ciphertext inflation (<1.5x), data is compressed using compression algorithms (such as LZ77) before encryption, and efficient encoding (such as integer encoding instead of floating-point numbers) is used in homomorphic encryption.

[0038] Step 5 aims to construct a fine-grained, identity-based long-term secure data storage and dynamic access control mechanism using the SM9 identity-based encryption algorithm. Its core lies in directly binding the decryption capability of encrypted data to the user's real identity, eliminating the need for pre-exchanging and managing traditional public key certificates. This is particularly suitable for scenarios in educational institutions where personnel roles (such as students, teachers, and administrators) change frequently. The specific steps are as follows: 1. System initialization and key generation Performed by a trusted key generation center (KGC) or the system's central server: (1) Generate SM9 master key pair: Perform SM9 algorithm initialization to generate system master public key. and the master private key Master key Publicly available, used for encryption; master private key Strictly confidential by KGC, used to generate user private keys.

[0039] (2) Define identity rules: Assign a unique identity to each user (or role) in the system. This identifier can be a structured string, such as "Organization Name:Department:Role:Unique ID" (e.g., "University_A:Math_Dept:Teacher:2021001").

[0040] 2. User private key distribution When a new user (such as a newly hired teacher) joins the system and needs to obtain data access permissions: (1) Identity authentication and submission: Users submit their identity identifiers to KGC through a secure channel. It provides trusted credentials (such as digital certificates, access cards, biometric information, etc.) to complete the binding and authentication of real identity and digital identifier.

[0041] (2) Generate user private key: KGC uses master private key and the user The user's private key is calculated using the key generation function in the SM9 algorithm. The calculation formula can be abstracted as follows: ; (3) Secure distribution of private keys: KGC distributes the private keys in a secure manner (such as using the SM2 algorithm to establish a secure channel for encrypted transmission). Distribute the private key to the user. The user must store and use the private key in a trusted environment (such as a hardware encryption module).

[0042] 3. Secure data storage (encrypted upload) Data owners (such as educational institutions) must encrypt data using SM9 before storing it in the cloud or on a central server. (1) Determine the access policy: The data owner specifies a list of user identities that are authorized to access this data. .

[0043] (2) Use SM9 to encrypt data: for sensitive data to be stored (This data may already be ciphertext encrypted in step 2, which is a second layer of access control encryption.) The data owner uses the SM9 master public key. Encrypt the specified set of identifiers using the SM9 encryption algorithm to generate ciphertext. The encryption process can be represented as: ; This cipher This means that only users whose identities are within the specified set can use their corresponding private keys. Only then can it be decrypted.

[0044] (3) Store ciphertext and metadata: Along with its hash value (For integrity verification) and access policy descriptions (i.e., a list of identities) are stored together in the security database as metadata.

[0045] 4. Dynamic access control (decrypted access) When authorized user When a request is made to access data within its permissions, the following process is triggered: (1) Access Request and Identity Statement: User Submit a data access request to the access control component, along with its identity identifier. .

[0046] (2) Identity verification: System verification The authenticity and validity of the identity (possibly in conjunction with a unified identity authentication system) are verified. After successful verification, the system retrieves stored metadata and checks it. Is it in the access policy list of this data file? If not available, access will be denied immediately.

[0047] (3) Generate and exchange decryption tokens (no interaction required or minimal interaction): Option A (decrypted by server calculation): The user decrypts it. Send to the server. The server utilizes the characteristics of the SM9 algorithm... Ciphertext with master key Perform partial decryption to generate a "conversion key" associated with the user's identity or partially decrypted intermediate ciphertext. It is then returned to the user. The user then uses their private key. right Perform the final decryption to obtain the plaintext. This process protects the user's private key from being exposed to the server.

[0048] Option B (completely decrypted by the client): If the data is encrypted If the access policy can be fully distributed to the client, then the user can directly use their own private key on the local client. Execute the SM9 decryption algorithm. Successful decryption requires the following conditions: Within the set of identities used during encryption.

[0049] (4) Decryption and result acquisition: The user terminal finally decrypts to obtain the original data. (or the previous layer of ciphertext). The user can then perform the integrity verification in step 4 (comparison) as needed. With stored hash value ) and subsequent decryption (such as decryption of the isomorphic encryption result).

[0050] 5. Permission revocation and update When it is necessary to cancel a user Access permissions for specific data: (1) Data re-encryption: The data owner or authorized administrator can access the policy list. Remove from Get a new list .

[0051] (2) Using the new identity list, process the original data through step 5.3. Re-encrypt to generate new ciphertext. .

[0052] (3) Update storage: using Replace old ciphertext in the database This will update the access policy in the metadata. Afterwards, users whose permissions have been revoked will be subject to penalties. If the new ciphertext is not in the authorized set, it cannot be decrypted.

[0053] (Optional) To improve efficiency, a proxy re-encryption or broadcast encryption variant based on SM9 can be used to reduce the re-encryption overhead for data owners.

[0054] like Figure 2 As shown, the present invention also provides a multimodal educational data security processing system based on national cryptographic algorithms and homomorphic encryption, used to implement the above-mentioned method, including: The preprocessing module is used to acquire multimodal educational data, classify it according to data type, and preprocess structured data, unstructured data, and time-series streaming data respectively to transform them into a unified format suitable for encrypted computing; The encryption module, connected to the preprocessing module, is used to generate an encryption key using the national cryptographic algorithm and to encrypt the preprocessed data using homomorphic encryption technology to generate the corresponding ciphertext. The computing module, which communicates with the encryption module, is used to perform homomorphic computation operations on ciphertext from multiple data sources in an encrypted state, thereby achieving privacy-preserving data analysis. The decryption and verification module communicates with the calculation module and is used by authorized users to decrypt the ciphertext of the calculation result using the corresponding private key, and to verify the integrity of the decryption result and the authenticity of the data source using the national cryptographic algorithm. The secure storage and access control module is connected to the encryption module and the decryption and verification module, respectively. It is used to manage data access permissions using the national cryptographic identity base encryption algorithm, securely store encrypted data, and implement identity-based access control.

[0055] Embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0056] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0057] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0058] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0059] Contents not described in detail in this specification are prior art known to those skilled in the art. It is hereby indicated that the above description is intended to help those skilled in the art understand this invention, but does not limit the scope of protection of this invention. Any equivalent substitutions, modifications, improvements, or simplifications of the above descriptions that do not depart from the essential content of this invention fall within the scope of protection of this invention.

Claims

1. A method for securely processing multimodal educational data based on national cryptographic algorithms and homomorphic encryption, characterized in that, include: Step 1: Acquire multimodal educational data, classify it according to data type, and preprocess each type of data to convert it into a unified format suitable for encrypted computing. The multimodal educational data includes structured data, unstructured data, and time-series streaming data. Step 2: Use the national cryptographic algorithm to generate an encryption key, and combine it with homomorphic encryption technology to encrypt the preprocessed data to generate the corresponding ciphertext; Step 3: In an encrypted state, perform homomorphic computation operations on ciphertext from multiple data sources to achieve privacy-preserving data analysis; Step 4: Authorize users to decrypt the ciphertext of the calculation result using the corresponding private key, and use the national cryptographic algorithm to verify the integrity of the decryption result and the authenticity of the data source; Step 5: Use the national cryptographic identity base encryption algorithm to manage data access permissions, securely store encrypted data, and implement identity-based access control.

2. The multimodal educational data security processing method based on national cryptographic algorithms and homomorphic encryption according to claim 1, characterized in that, In step 1, the structured data is preprocessed, specifically as follows: Extract the key fields from the structured data and serialize them into binary format.

3. The multimodal educational data security processing method based on national cryptographic algorithms and homomorphic encryption according to claim 2, characterized in that, In step 1, the unstructured data is preprocessed, specifically as follows: Unstructured data is processed in segments and features are extracted. The feature extraction is implemented using a convolutional neural network.

4. The multimodal educational data security processing method based on national cryptographic algorithms and homomorphic encryption according to claim 3, characterized in that, In step 1, the time-series stream data is preprocessed, specifically as follows: The time-series stream data is subjected to window segmentation and normalization processing, wherein the window segmentation is based on time window sampling.

5. The multimodal educational data security processing method based on national cryptographic algorithms and homomorphic encryption according to claim 4, characterized in that, In step 2, the encryption key is generated using the national cryptographic algorithm, specifically as follows: Asymmetric key pair public key generated based on SM2 algorithm PK sm2 and private key SK sm2 Used for digital signatures and key exchange; Calculate the hash value of the data based on the SM3 algorithm: H = SM 3 ( D (), where D represents the original data and H represents the hash value, used for integrity verification; Generate a symmetric key based on the SM4 algorithm K sm4 It is used to encrypt unstructured data and time-series streaming data; Generate the identity base key using the SM9 algorithm: Public Key PK sm9 and private key SK sm9 Used for user-identity-based secure access control; Generate homomorphic encryption key pair: public key PK he and private key SK he Homomorphic encryption schemes employ either the Paillier algorithm or the BFV scheme.

6. The multimodal educational data security processing method based on national cryptographic algorithms and homomorphic encryption according to claim 5, characterized in that, In step 2, homomorphic encryption is used to encrypt the preprocessed data, generating the corresponding ciphertext, specifically as follows: For structured data, use public keys. PK he Encrypt; For unstructured data, first use a symmetric key. K sm4 Encrypt it, then use the public key. PK he For symmetric keys K sm4 Encrypt; For time-series streaming data, first use the public key. PK he Perform streaming encryption, encrypting each normalized data point individually.

7. The multimodal educational data security processing method based on national cryptographic algorithms and homomorphic encryption according to claim 6, characterized in that, In step 3, the homomorphic computation operations include: homomorphic addition operations, homomorphic multiplication operations, and federated statistical process operations.

8. The multimodal educational data security processing method based on national cryptographic algorithms and homomorphic encryption according to claim 7, characterized in that, In step 4, the authorized user decrypts the ciphertext of the calculation result using the corresponding private key, and verifies the integrity of the decryption result and the authenticity of the data source using the national cryptographic algorithm. Specifically: Authorized users can use the corresponding private key to decrypt the ciphertext; The hash value of the decrypted data is calculated based on the SM3 algorithm and compared with the original hash value. If they are the same, the data is considered complete. Based on public key PK sm2 Verify the authenticity of the decrypted data.

9. The multimodal educational data security processing method based on national cryptographic algorithms and homomorphic encryption according to claim 8, characterized in that, In step 5, the national cryptographic identity base encryption algorithm is used to manage data access permissions, specifically as follows: Data access permissions are managed based on the SM9 algorithm. When a user requests data, the user's identity is verified using the SM9 private key. SK sm9 Generate a decryption token. If the token is valid, allow the user to access the data; otherwise, do not allow access.

10. A multimodal educational data security processing system based on national cryptographic algorithms and homomorphic encryption, used to implement the method of any one of claims 1-9, characterized in that, include: The preprocessing module is used to acquire multimodal educational data, classify it according to data type, and preprocess structured data, unstructured data, and time-series streaming data respectively to transform them into a unified format suitable for encrypted computing; The encryption module, connected to the preprocessing module, is used to generate an encryption key using the national cryptographic algorithm and to encrypt the preprocessed data using homomorphic encryption technology to generate the corresponding ciphertext. The computing module, which communicates with the encryption module, is used to perform homomorphic computation operations on ciphertext from multiple data sources in an encrypted state, thereby achieving privacy-preserving data analysis. The decryption and verification module communicates with the calculation module and is used by authorized users to decrypt the ciphertext of the calculation result using the corresponding private key, and to verify the integrity of the decryption result and the authenticity of the data source using the national cryptographic algorithm. The secure storage and access control module is connected to the encryption module and the decryption and verification module, respectively. It is used to manage data access permissions using the national cryptographic identity base encryption algorithm, securely store encrypted data, and implement identity-based access control.

Images