Convolutional neural network model over encrypted data based on functional encryption
Currently, homomorphic encryption, secure multi-party computation, and other encryption schemes are used to protect the privacy of sensitive data in outsourced convolutional neural network (CNN) models.However, the computational and communication overhead caused by the above schemes would reduce sys...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial Department of Journal on Communications
2024-03-01
|
| Series: | Tongxin xuebao |
| Subjects: | |
| Online Access: | http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2024050/ |
| _version_ | 1826875442976522240 |
|---|---|
| author | Chen WANG Jiarun LI Jian XU |
| author_facet | Chen WANG Jiarun LI Jian XU |
| author_sort | Chen WANG |
| collection | DOAJ |
| description | Currently, homomorphic encryption, secure multi-party computation, and other encryption schemes are used to protect the privacy of sensitive data in outsourced convolutional neural network (CNN) models.However, the computational and communication overhead caused by the above schemes would reduce system efficiency.Based on the low cost of functional encryption, a new convolutional neural network model over encrypted data was constructed using functional encryption.Firstly, two algorithms based on functional encryption were designed, including inner product functional encryption and basic operation functional encryption algorithms to implement basic operations such as inner product, multiplication, and subtraction over encrypted data, reducing computational and communication costs.Secondly, a secure convolutional computation protocol and a secure loss optimization protocol were designed for each of these basic operations, which achieved ciphertext forward propagation in the convolutional layer and ciphertext backward propagation in the output layer.Finally, a secure training and classification method for the model was provided by the above secure protocols in a module-composable way, which could simultaneously protect the confidentiality of user data as well as data labels.Theoretical analysis and experimental results indicate that the proposed model can achieve CNN training and classification over encrypted data while ensuring accuracy and security. |
| first_indexed | 2025-02-17T00:49:30Z |
| format | Article |
| id | doaj.art-4cf3378dc83f4ce98b765d393283cf49 |
| institution | Directory Open Access Journal |
| issn | 1000-436X |
| language | zho |
| last_indexed | 2025-02-17T00:49:30Z |
| publishDate | 2024-03-01 |
| publisher | Editorial Department of Journal on Communications |
| record_format | Article |
| series | Tongxin xuebao |
| spelling | doaj.art-4cf3378dc83f4ce98b765d393283cf492025-01-14T06:21:51ZzhoEditorial Department of Journal on CommunicationsTongxin xuebao1000-436X2024-03-0145506559296350Convolutional neural network model over encrypted data based on functional encryptionChen WANGJiarun LIJian XUCurrently, homomorphic encryption, secure multi-party computation, and other encryption schemes are used to protect the privacy of sensitive data in outsourced convolutional neural network (CNN) models.However, the computational and communication overhead caused by the above schemes would reduce system efficiency.Based on the low cost of functional encryption, a new convolutional neural network model over encrypted data was constructed using functional encryption.Firstly, two algorithms based on functional encryption were designed, including inner product functional encryption and basic operation functional encryption algorithms to implement basic operations such as inner product, multiplication, and subtraction over encrypted data, reducing computational and communication costs.Secondly, a secure convolutional computation protocol and a secure loss optimization protocol were designed for each of these basic operations, which achieved ciphertext forward propagation in the convolutional layer and ciphertext backward propagation in the output layer.Finally, a secure training and classification method for the model was provided by the above secure protocols in a module-composable way, which could simultaneously protect the confidentiality of user data as well as data labels.Theoretical analysis and experimental results indicate that the proposed model can achieve CNN training and classification over encrypted data while ensuring accuracy and security.http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2024050/convolutional neural networkencrypted datafunctional encryptionprivacy protection |
| spellingShingle | Chen WANG Jiarun LI Jian XU Convolutional neural network model over encrypted data based on functional encryption Tongxin xuebao convolutional neural network encrypted data functional encryption privacy protection |
| title | Convolutional neural network model over encrypted data based on functional encryption |
| title_full | Convolutional neural network model over encrypted data based on functional encryption |
| title_fullStr | Convolutional neural network model over encrypted data based on functional encryption |
| title_full_unstemmed | Convolutional neural network model over encrypted data based on functional encryption |
| title_short | Convolutional neural network model over encrypted data based on functional encryption |
| title_sort | convolutional neural network model over encrypted data based on functional encryption |
| topic | convolutional neural network encrypted data functional encryption privacy protection |
| url | http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2024050/ |
| work_keys_str_mv | AT chenwang convolutionalneuralnetworkmodeloverencrypteddatabasedonfunctionalencryption AT jiarunli convolutionalneuralnetworkmodeloverencrypteddatabasedonfunctionalencryption AT jianxu convolutionalneuralnetworkmodeloverencrypteddatabasedonfunctionalencryption |